
# importing libraries required for the project
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import numpy as np  # linear algebra
import matplotlib.pyplot as plt # for plotting
import seaborn as sns   # for plotting
import warnings # ignore warnings
warnings.filterwarnings('ignore') 
from sklearn.preprocessing import StandardScaler,MinMaxScaler # for scaling
from sklearn.model_selection import train_test_split # for splitting the data

# Machine Learning Libraries
from sklearn.linear_model import LogisticRegression 

# Model Evaluation Libraries
from sklearn.metrics import confusion_matrix,recall_score,  roc_auc_score, roc_curve,f1_score
from sklearn.metrics import accuracy_score,precision_score, classification_report,precision_recall_curve
# Oversampling Libraries
from imblearn.over_sampling import SMOTE,ADASYN,BorderlineSMOTE,KMeansSMOTE,SMOTENC,SVMSMOTE

from sklearn.tree import DecisionTreeClassifier
from xgboost import XGBClassifier
from catboost import CatBoostClassifier
from sklearn.ensemble import RandomForestClassifier,GradientBoostingClassifier,AdaBoostClassifier
from sklearn.neighbors import KNeighborsClassifier
from sklearn.model_selection import GridSearchCV,RandomizedSearchCV
from lightgbm import LGBMClassifier
from hyperopt import hp,fmin,tpe,STATUS_OK,Trials
from sklearn.model_selection import cross_val_score


# reading the csv file from pandas library
bank_data=pd.read_csv('COMPANY BANKRUPTCY PREDICTION.csv') 

# setting the display option to show all columns
pd.set_option('display.max_columns', None)
# setting the display option to show 100 rows
pd.set_option('display.max_rows', 100)


# Print the first 5 rows of the dataframe
bank_data.head()


# printing the tail of the dataframe
bank_data.tail()


# Checking value counts of the target variable
target_counts=bank_data["Bankrupt?"].value_counts()
print(target_counts) 
print(np.round(target_counts[1]/target_counts[0]),3)

0    6599
1     220
Name: Bankrupt?, dtype: int64
0.0 3


sns.countplot(bank_data['Bankrupt?']) # plotting the count plot of target variable

<AxesSubplot:xlabel='Bankrupt?', ylabel='count'>


bank_data.info() # to know the data types of the columns

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 6819 entries, 0 to 6818
Data columns (total 96 columns):
 #   Column                                                    Non-Null Count  Dtype  
---  ------                                                    --------------  -----  
 0   Bankrupt?                                                 6819 non-null   int64  
 1    ROA(C) before interest and depreciation before interest  6819 non-null   float64
 2    ROA(A) before interest and % after tax                   6819 non-null   float64
 3    ROA(B) before interest and depreciation after tax        6819 non-null   float64
 4    Operating Gross Margin                                   6819 non-null   float64
 5    Realized Sales Gross Margin                              6819 non-null   float64
 6    Operating Profit Rate                                    6819 non-null   float64
 7    Pre-tax net Interest Rate                                6819 non-null   float64
 8    After-tax net Interest Rate                              6819 non-null   float64
 9    Non-industry income and expenditure/revenue              6819 non-null   float64
 10   Continuous interest rate (after tax)                     6819 non-null   float64
 11   Operating Expense Rate                                   6819 non-null   float64
 12   Research and development expense rate                    6819 non-null   float64
 13   Cash flow rate                                           6819 non-null   float64
 14   Interest-bearing debt interest rate                      6819 non-null   float64
 15   Tax rate (A)                                             6819 non-null   float64
 16   Net Value Per Share (B)                                  6819 non-null   float64
 17   Net Value Per Share (A)                                  6819 non-null   float64
 18   Net Value Per Share (C)                                  6819 non-null   float64
 19   Persistent EPS in the Last Four Seasons                  6819 non-null   float64
 20   Cash Flow Per Share                                      6819 non-null   float64
 21   Revenue Per Share (Yuan ¥)                               6819 non-null   float64
 22   Operating Profit Per Share (Yuan ¥)                      6819 non-null   float64
 23   Per Share Net profit before tax (Yuan ¥)                 6819 non-null   float64
 24   Realized Sales Gross Profit Growth Rate                  6819 non-null   float64
 25   Operating Profit Growth Rate                             6819 non-null   float64
 26   After-tax Net Profit Growth Rate                         6819 non-null   float64
 27   Regular Net Profit Growth Rate                           6819 non-null   float64
 28   Continuous Net Profit Growth Rate                        6819 non-null   float64
 29   Total Asset Growth Rate                                  6819 non-null   float64
 30   Net Value Growth Rate                                    6819 non-null   float64
 31   Total Asset Return Growth Rate Ratio                     6819 non-null   float64
 32   Cash Reinvestment %                                      6819 non-null   float64
 33   Current Ratio                                            6819 non-null   float64
 34   Quick Ratio                                              6819 non-null   float64
 35   Interest Expense Ratio                                   6819 non-null   float64
 36   Total debt/Total net worth                               6819 non-null   float64
 37   Debt ratio %                                             6819 non-null   float64
 38   Net worth/Assets                                         6819 non-null   float64
 39   Long-term fund suitability ratio (A)                     6819 non-null   float64
 40   Borrowing dependency                                     6819 non-null   float64
 41   Contingent liabilities/Net worth                         6819 non-null   float64
 42   Operating profit/Paid-in capital                         6819 non-null   float64
 43   Net profit before tax/Paid-in capital                    6819 non-null   float64
 44   Inventory and accounts receivable/Net value              6819 non-null   float64
 45   Total Asset Turnover                                     6819 non-null   float64
 46   Accounts Receivable Turnover                             6819 non-null   float64
 47   Average Collection Days                                  6819 non-null   float64
 48   Inventory Turnover Rate (times)                          6819 non-null   float64
 49   Fixed Assets Turnover Frequency                          6819 non-null   float64
 50   Net Worth Turnover Rate (times)                          6819 non-null   float64
 51   Revenue per person                                       6819 non-null   float64
 52   Operating profit per person                              6819 non-null   float64
 53   Allocation rate per person                               6819 non-null   float64
 54   Working Capital to Total Assets                          6819 non-null   float64
 55   Quick Assets/Total Assets                                6819 non-null   float64
 56   Current Assets/Total Assets                              6819 non-null   float64
 57   Cash/Total Assets                                        6819 non-null   float64
 58   Quick Assets/Current Liability                           6819 non-null   float64
 59   Cash/Current Liability                                   6819 non-null   float64
 60   Current Liability to Assets                              6819 non-null   float64
 61   Operating Funds to Liability                             6819 non-null   float64
 62   Inventory/Working Capital                                6819 non-null   float64
 63   Inventory/Current Liability                              6819 non-null   float64
 64   Current Liabilities/Liability                            6819 non-null   float64
 65   Working Capital/Equity                                   6819 non-null   float64
 66   Current Liabilities/Equity                               6819 non-null   float64
 67   Long-term Liability to Current Assets                    6819 non-null   float64
 68   Retained Earnings to Total Assets                        6819 non-null   float64
 69   Total income/Total expense                               6819 non-null   float64
 70   Total expense/Assets                                     6819 non-null   float64
 71   Current Asset Turnover Rate                              6819 non-null   float64
 72   Quick Asset Turnover Rate                                6819 non-null   float64
 73   Working capitcal Turnover Rate                           6819 non-null   float64
 74   Cash Turnover Rate                                       6819 non-null   float64
 75   Cash Flow to Sales                                       6819 non-null   float64
 76   Fixed Assets to Assets                                   6819 non-null   float64
 77   Current Liability to Liability                           6819 non-null   float64
 78   Current Liability to Equity                              6819 non-null   float64
 79   Equity to Long-term Liability                            6819 non-null   float64
 80   Cash Flow to Total Assets                                6819 non-null   float64
 81   Cash Flow to Liability                                   6819 non-null   float64
 82   CFO to Assets                                            6819 non-null   float64
 83   Cash Flow to Equity                                      6819 non-null   float64
 84   Current Liability to Current Assets                      6819 non-null   float64
 85   Liability-Assets Flag                                    6819 non-null   int64  
 86   Net Income to Total Assets                               6819 non-null   float64
 87   Total assets to GNP price                                6819 non-null   float64
 88   No-credit Interval                                       6819 non-null   float64
 89   Gross Profit to Sales                                    6819 non-null   float64
 90   Net Income to Stockholder's Equity                       6819 non-null   float64
 91   Liability to Equity                                      6819 non-null   float64
 92   Degree of Financial Leverage (DFL)                       6819 non-null   float64
 93   Interest Coverage Ratio (Interest expense to EBIT)       6819 non-null   float64
 94   Net Income Flag                                          6819 non-null   int64  
 95   Equity to Liability                                      6819 non-null   float64
dtypes: float64(93), int64(3)
memory usage: 5.0 MB


bank_data.shape # printing the shape of the dataframe

(6819, 96)


bank_data.describe()# to know the statistical summary of the dataframe


for i in  bank_data.describe().columns:
    # print(len(bank_data[i].unique()) )
    if len(bank_data[i].unique()) < 10:
        print(i,'with',len(bank_data[i].unique()),'values')

Bankrupt? with 2 values
 Liability-Assets Flag with 2 values
 Net Income Flag with 1 values


print(bank_data[' Net Income Flag'].unique())
# Drop the columns Net Income Flag which is not required
bank_data.drop([' Net Income Flag'],axis=1,inplace=True)

[1]


pd.DataFrame(data=bank_data.isna().mean()*100,index=bank_data.columns).T


sns.heatmap(bank_data.isnull(), yticklabels=False, cbar=False, cmap='coolwarm')

<AxesSubplot:>


bank_data.isna().sum().sum()

0


numerical_columns=[x for x in bank_data.columns if x not in [' Net Income Flag','Bankrupt?'] ]
numerical_columns #printing the numerical columns

[' ROA(C) before interest and depreciation before interest',
 ' ROA(A) before interest and % after tax',
 ' ROA(B) before interest and depreciation after tax',
 ' Operating Gross Margin',
 ' Realized Sales Gross Margin',
 ' Operating Profit Rate',
 ' Pre-tax net Interest Rate',
 ' After-tax net Interest Rate',
 ' Non-industry income and expenditure/revenue',
 ' Continuous interest rate (after tax)',
 ' Operating Expense Rate',
 ' Research and development expense rate',
 ' Cash flow rate',
 ' Interest-bearing debt interest rate',
 ' Tax rate (A)',
 ' Net Value Per Share (B)',
 ' Net Value Per Share (A)',
 ' Net Value Per Share (C)',
 ' Persistent EPS in the Last Four Seasons',
 ' Cash Flow Per Share',
 ' Revenue Per Share (Yuan ¥)',
 ' Operating Profit Per Share (Yuan ¥)',
 ' Per Share Net profit before tax (Yuan ¥)',
 ' Realized Sales Gross Profit Growth Rate',
 ' Operating Profit Growth Rate',
 ' After-tax Net Profit Growth Rate',
 ' Regular Net Profit Growth Rate',
 ' Continuous Net Profit Growth Rate',
 ' Total Asset Growth Rate',
 ' Net Value Growth Rate',
 ' Total Asset Return Growth Rate Ratio',
 ' Cash Reinvestment %',
 ' Current Ratio',
 ' Quick Ratio',
 ' Interest Expense Ratio',
 ' Total debt/Total net worth',
 ' Debt ratio %',
 ' Net worth/Assets',
 ' Long-term fund suitability ratio (A)',
 ' Borrowing dependency',
 ' Contingent liabilities/Net worth',
 ' Operating profit/Paid-in capital',
 ' Net profit before tax/Paid-in capital',
 ' Inventory and accounts receivable/Net value',
 ' Total Asset Turnover',
 ' Accounts Receivable Turnover',
 ' Average Collection Days',
 ' Inventory Turnover Rate (times)',
 ' Fixed Assets Turnover Frequency',
 ' Net Worth Turnover Rate (times)',
 ' Revenue per person',
 ' Operating profit per person',
 ' Allocation rate per person',
 ' Working Capital to Total Assets',
 ' Quick Assets/Total Assets',
 ' Current Assets/Total Assets',
 ' Cash/Total Assets',
 ' Quick Assets/Current Liability',
 ' Cash/Current Liability',
 ' Current Liability to Assets',
 ' Operating Funds to Liability',
 ' Inventory/Working Capital',
 ' Inventory/Current Liability',
 ' Current Liabilities/Liability',
 ' Working Capital/Equity',
 ' Current Liabilities/Equity',
 ' Long-term Liability to Current Assets',
 ' Retained Earnings to Total Assets',
 ' Total income/Total expense',
 ' Total expense/Assets',
 ' Current Asset Turnover Rate',
 ' Quick Asset Turnover Rate',
 ' Working capitcal Turnover Rate',
 ' Cash Turnover Rate',
 ' Cash Flow to Sales',
 ' Fixed Assets to Assets',
 ' Current Liability to Liability',
 ' Current Liability to Equity',
 ' Equity to Long-term Liability',
 ' Cash Flow to Total Assets',
 ' Cash Flow to Liability',
 ' CFO to Assets',
 ' Cash Flow to Equity',
 ' Current Liability to Current Assets',
 ' Liability-Assets Flag',
 ' Net Income to Total Assets',
 ' Total assets to GNP price',
 ' No-credit Interval',
 ' Gross Profit to Sales',
 " Net Income to Stockholder's Equity",
 ' Liability to Equity',
 ' Degree of Financial Leverage (DFL)',
 ' Interest Coverage Ratio (Interest expense to EBIT)',
 ' Equity to Liability']


from IPython.display import display # to display the dataframe 

def display_quatile_dist(data,features):
    ''' 
    data: dataframe
    features: list of features
    This function is used to display the distribution of the features in the dataframe
    from the list of percentile values {0.0,0.01,0.1,0.25,0.5,0.75,0.9,0.99,1}
    displays distribution of the features in the dataframe through the list of percentile values
    '''
    a=pd.DataFrame()
    r=[0,.01,.1,.25,.5,.75,.9,.99,1]
    for i in features :
#     display(bank_data[i].quantile(r))
        a[i]=data[i].quantile(r)
    
    display(a.style.bar())
    
display_quatile_dist(bank_data,numerical_columns)


# Setting the size of the plot
plt.figure(figsize=(12,12),dpi=200)
sns.heatmap(bank_data.corr(),) # heatmap of the correlation between the features

<AxesSubplot:>


def correlation(dataset, threshold):
    '''
    dataset: dataframe
    threshold: threshold value for the correlation
    Type:dataset: dataframe
    Type:threshold: float
    This function removes the features with correlation greater than the threshold value
    and it alters the dataframe to remove the features with correlation greater than the threshold value
    '''
    col_corr = set() # Set of all the names of deleted columns
    corr_matrix = dataset.corr() # Correlation matrix
    # Iterating through the correlation matrix
    for i in range(len(corr_matrix.columns)):
        for j in range(i):
            if (corr_matrix.iloc[i, j] >= threshold) and (corr_matrix.columns[j] not in col_corr):
                colname = corr_matrix.columns[i] # getting the name of column
                col_corr.add(colname)
                if colname in dataset.columns:
                    del dataset[colname] # deleting the column from the dataset


# Setting a copy of the dataset
bank_data_stage_1=bank_data.copy()
correlation(bank_data_stage_1,.85)


high_corr_cols=set(bank_data.columns)-set(bank_data_stage_1.columns)
plt.figure(figsize=(12,12),dpi=200)
sns.heatmap(bank_data[high_corr_cols].corr(),annot=True)

<AxesSubplot:>


display(bank_data_stage_1.head()) # printing the first 5 rows of the dataset
print('Printing shape of the dataset after removing the features with high correlation to check the features left')
bank_data_stage_1.shape # printing the shape of the dataset

Printing shape of the dataset after removing the features with high correlation to check the features left

(6819, 74)


(display_quatile_dist(bank_data_stage_1,bank_data_stage_1.columns))


# storing features in a list
features=[x for x in bank_data_stage_1.columns if x not in ["Bankrupt?"]]


def graph_label_title(x_label=False,y_label=False,title=False):
    '''
    x_label: label of the x-axis
    y_label: label of the y-axis
    title: title of the graph
    This function sets the xlable ylable and title of the graph
    '''
    if x_label:
        plt.xlabel(x_label,fontsize=25)
    if y_label:
        plt.ylabel(y_label,fontsize=25)
    if title:
        plt.title(title,fontsize=25)


# Plottiong the distribution of skewness of features
skew_table=bank_data_stage_1.drop(columns=['Bankrupt?'," Liability-Assets Flag"],axis=1).skew().sort_values()
print( "\033[1m" + 'Top 5 features that skewed negatively' + "\033[0m")
display(pd.DataFrame(skew_table.head())) # printing the top 5 features that skewed negatively
print( "\033[1m" + 'Top 5 features that skewed positively' + "\033[0m")
display(pd.DataFrame(skew_table.tail())) # printing the top 5 features that skewed positively

Top 5 features that skewed negatively

Top 5 features that skewed positively


# Plotting the distribution of skewness of features
plt.figure(figsize=(5,3),dpi=200) # setting the size of the plot
sns.distplot(skew_table)
graph_label_title('skewed distributions',False,'Distributions of Skewness of features')


bank_data["Bankrupt?"].value_counts() # printing the value counts of the target variable

0    6599
1     220
Name: Bankrupt?, dtype: int64


def print_skewness_and_handle_outlier(skewness_range_1,skewness_range_2,data,skewness,percentile=.99,inplace=False):
    '''
    skewness_range_1: range 1 for skewness
    skewness_range_2: range 2 for skewness
    data: dataframe
    skewness: indication that its a negative or positive skewed feature
    percentile: percentile value to eliminate the outlier
    inplace: indication that the dataframe is to be altered or not

    TYPE:
    skewness_range_1: int
    skewness_range_2: int
    data: dataframe
    skewness: string
    percentile: float
    inplace: boolean

    This function prints extracts features between the range of skewness and eliminates the outlier
    using functions that are
    1. return_and_print_outliers_ : 
    2.skewness range : To get the features that are between the range of skewness

    Presentations of the function
        BEFORE removing Outliers AND AFTER removing Outliers 
        1. It displays first 5 rows of the dataframe with features between the skewness range
        2. Plots the box-plot of the features between the skewness range
        3. Display the percentile distrubution
        4. Histogram to check the distribution of the features or the how much the skewness is reduced by removing the outliers

    '''
    data_=data.copy()
    skew_cols=skewness_Range(skewness_range_1,skewness_range_2)
    print("*"*4+f'Displaying first 5 rows of features that are having it skewness from {skewness_range_1} to {skewness_range_2}'+"*"*4)
    display(data[skew_cols].head())
    
    print("*"*4+' Box plot before removal of outliers'+"*"*4)
    plt.figure(figsize=(15,3))
    ax = sns.boxplot(data=data_[skew_cols], orient="h", palette="Set2")
    plt.show()
    
    print('\n'+"*"*4+f' Displaying percentile of columns that are skewed from {skewness_range_1} to {skewness_range_2} '+"*"*4)
    print('\n'+'*'*55+"BEFORE"+'*'*55)
    display_quatile_dist(data_,skew_cols)
    plt.show()
    print("\nSKEWNESS")
    print('\n'+'*'*55+"BEFORE"+'*'*55)
    data_[skew_cols].hist(bins=50,figsize=(12,8))
    plt.show()    
    outlier_indices=return_and_print_outliers_(data,skew_cols,percentile,skewness,inplace)
    data.drop(outlier_indices[0],inplace=inplace)
    return skew_cols


def return_and_print_outliers_(data,cols,percentile,skewness,inplace):
    '''
    data: dataframe 
    cols: list of columns
    percentile: percentile value to eliminate the outlier
    skewness: indication that its a negative or positive skewed feature
    inplace: indication that the dataframe is to be altered or not
    TYPE :
    data: dataframe
    cols: list
    percentile: float
    skewness: string
    inplace: boolean

    1.This function identifies data above supplied percentile and impute
    2.displays the graphs like histogram boxplot and percentile distrubution chart after removing outliers
    This function returns the indices of the majority and minority class that were imputed due the reason of outlier
    '''
    print(f'\n\n {"*"*10}Outliers beyond {percentile} percentile {"*"*10}')
    # set used to collected majority class outlier indices
    outliers_indcies=set()
    # set used to collected minority class outlier indices
    outliers_indcies_minority=set()
    
    # dataframe to store the outliers from majority class
    a=pd.DataFrame()
    # dataframe to store the outliers from minority class
    b=pd.DataFrame()
    # Iterating over the columns supplied
    for i in cols:
        # condition for skewness is + i.e positive eg: we need to supply percentile like .99 ..95 .90 percentile
        if skewness=="+":
            # storing data that are above percentile
            outlier_=data[(data[i]> percentile)]
            # storing  the outliers belonging to majority class
            outlier=outlier_[(outlier_["Bankrupt?"]!=1)]
            # storing  the outliers belonging to minority class
            outlier_minority=outlier_[(outlier_["Bankrupt?"]!=0)]
            # condition to check if there are outliers captured and inplace value [Bool] to alter the origiial dataframe
            if data.loc[(data[i]> percentile) &  (data["Bankrupt?"]!=0),[i]].shape[0] > 0 and inplace:
                # imputing overall median for minority class outliers
                data.loc[(data[i]> percentile) &  (data["Bankrupt?"]!=0),[i]]==data[i].median()
        # Condition for skewness is - i.e negative eg: we need to supply percentile like .01 ..05 .10 percentile
        elif skewness=='-': 
            # storing data that are below percentile
            outlier_=data[(data[i] < percentile)]
            # storing  the outliers belonging to majority class
            outlier=outlier_[(outlier_["Bankrupt?"]!=1)]
            # storing  the outliers belonging to minority class
            outlier_minority=outlier_[(outlier_["Bankrupt?"]!=0)]
            # condition to check if there are outliers captured and inplace value [Bool] to alter the origiial dataframe
            if data.loc[(data[i]  < percentile) &  (data["Bankrupt?"]!=0),[i]].shape[0] > 0 and inplace:
                # imputing overall median for minority class outliers
                data.loc[(data[i] < percentile) &  (data["Bankrupt?"]!=0),[i]]=data[i].median()
        # Storing indices of the outliers captured for majority class
        outliers_indcies.update(outlier.index)
        # Storing indices of the outliers captured for minority class
        outliers_indcies_minority.update(outlier_minority.index)
        # Concatenating the outliers captured for majority class
        a=pd.concat([a,outlier])
        # Concatenating the outliers captured for minority class
        b=pd.concat([b,outlier_minority])
   
    # Displaying the outliers captured for majority class and minority class
    if a[cols].shape[1]!=0:
        modified_data=pd.concat([a,b]).loc[list(outliers_indcies)+list(outliers_indcies_minority),cols+["Bankrupt?"]]
        
        try:
            display(modified_data.reset_index(drop=True).style.bar())
#             pass
        except:
            display(modified_data.reset_index())
        print(len(outliers_indcies_minority),'minority class')
            

# Presentation Part
    print(f'{b.shape[0]} rows of specific column  are imputed to median value ')
    print(f'{a.shape[0]} rows are removed ')
    print("\nSKEWNESS")
    print('\n'+'*'*55+"AFTER"+'*'*55)
    data.drop(list(outliers_indcies)+list(outliers_indcies_minority))[cols].hist(bins=50,figsize=(12,8))
    plt.show()     
    print("*"*4+' Box plot after removal of outliers'+"*"*4)
    plt.figure(figsize=(15,3))
    sns.boxplot(data=data[cols].drop(list(outliers_indcies)+list(outliers_indcies_minority)), orient="h", palette="Set2")
    plt.show()
    print("\nPERCENTILE")
    print('*'*55+"AFTER"+'*'*55)
    display_quatile_dist(data[cols].drop(list(outliers_indcies)+list(outliers_indcies_minority)),cols)
    plt.show()
    
    return [outliers_indcies,outliers_indcies_minority]

def skewness_Range(range1,range2):
    '''
    range1: range1 of the skew
    range2: range2 of the skew
    TYPE:
    range1: int or float
    range2: int or float

    This function returns the feature that having its skewnes between the  skew range
    '''
    cols_in_range=[]
    for ind in skew_table.index:
        if range2 > skew_table[ind] >= range1 :
            cols_in_range.append(ind)
    return cols_in_range


# 70 amd 85 is the range of the skewness it selects the features between that range
# '+' indicates that we have checking positively skewed features
# .90 is the percentile threshold supplied 
print_skewness_and_handle_outlier(70,85,bank_data_stage_1,'+',.90,True)

****Displaying first 5 rows of features that are having it skewness from 70 to 85****

**** Box plot before removal of outliers****

**** Displaying percentile of columns that are skewed from 70 to 85 ****

*******************************************************BEFORE*******************************************************

SKEWNESS

*******************************************************BEFORE*******************************************************


 **********Outliers beyond 0.9 percentile **********

4 minority class
4 rows of specific column  are imputed to median value 
15 rows are removed 

SKEWNESS

*******************************************************AFTER*******************************************************

**** Box plot after removal of outliers****

PERCENTILE
*******************************************************AFTER*******************************************************

[' Realized Sales Gross Profit Growth Rate',
 ' Contingent liabilities/Net worth',
 ' Net Value Growth Rate',
 ' Total income/Total expense',
 ' Current Ratio',
 ' Fixed Assets to Assets']


print_skewness_and_handle_outlier(60,70,bank_data_stage_1,'+',.8,True)

****Displaying first 5 rows of features that are having it skewness from 60 to 70****

**** Box plot before removal of outliers****

**** Displaying percentile of columns that are skewed from 60 to 70 ****

*******************************************************BEFORE*******************************************************

SKEWNESS

*******************************************************BEFORE*******************************************************


 **********Outliers beyond 0.8 percentile **********

0 minority class
0 rows of specific column  are imputed to median value 
2 rows are removed 

SKEWNESS

*******************************************************AFTER*******************************************************

**** Box plot after removal of outliers****

PERCENTILE
*******************************************************AFTER*******************************************************

[' Total Asset Return Growth Rate Ratio', ' Continuous Net Profit Growth Rate']


print_skewness_and_handle_outlier(45,61,bank_data_stage_1,'+',.7,True)

****Displaying first 5 rows of features that are having it skewness from 45 to 61****

**** Box plot before removal of outliers****

**** Displaying percentile of columns that are skewed from 45 to 61 ****

*******************************************************BEFORE*******************************************************

SKEWNESS

*******************************************************BEFORE*******************************************************


 **********Outliers beyond 0.7 percentile **********

3 minority class
3 rows of specific column  are imputed to median value 
13 rows are removed 

SKEWNESS

*******************************************************AFTER*******************************************************

**** Box plot after removal of outliers****

PERCENTILE
*******************************************************AFTER*******************************************************

[' Inventory/Working Capital',
 ' Degree of Financial Leverage (DFL)',
 ' Total debt/Total net worth',
 ' Quick Assets/Current Liability',
 ' Revenue per person']


print_skewness_and_handle_outlier(30,44,bank_data_stage_1,'+',.8,True)

****Displaying first 5 rows of features that are having it skewness from 30 to 44****

**** Box plot before removal of outliers****

**** Displaying percentile of columns that are skewed from 30 to 44 ****

*******************************************************BEFORE*******************************************************

SKEWNESS

*******************************************************BEFORE*******************************************************


 **********Outliers beyond 0.8 percentile **********

4 minority class
4 rows of specific column  are imputed to median value 
31 rows are removed 

SKEWNESS

*******************************************************AFTER*******************************************************

**** Box plot after removal of outliers****

PERCENTILE
*******************************************************AFTER*******************************************************

[' Average Collection Days',
 ' Quick Ratio',
 ' Equity to Long-term Liability',
 ' Non-industry income and expenditure/revenue',
 ' Revenue Per Share (Yuan ¥)']


print_skewness_and_handle_outlier(20,30,bank_data_stage_1,'+',.9,True)

****Displaying first 5 rows of features that are having it skewness from 20 to 30****

**** Box plot before removal of outliers****

**** Displaying percentile of columns that are skewed from 20 to 30 ****

*******************************************************BEFORE*******************************************************

SKEWNESS

*******************************************************BEFORE*******************************************************


 **********Outliers beyond 0.9 percentile **********

7 minority class
8 rows of specific column  are imputed to median value 
46 rows are removed 

SKEWNESS

*******************************************************AFTER*******************************************************

**** Box plot after removal of outliers****

PERCENTILE
*******************************************************AFTER*******************************************************

[' Borrowing dependency',
 ' Total assets to GNP price',
 ' Long-term fund suitability ratio (A)',
 ' Accounts Receivable Turnover',
 ' Allocation rate per person']


print_skewness_and_handle_outlier(13,20,bank_data_stage_1,'+',.95,True)

****Displaying first 5 rows of features that are having it skewness from 13 to 20****

**** Box plot before removal of outliers****

**** Displaying percentile of columns that are skewed from 13 to 20 ****

*******************************************************BEFORE*******************************************************

SKEWNESS

*******************************************************BEFORE*******************************************************


 **********Outliers beyond 0.95 percentile **********

15 minority class
15 rows of specific column  are imputed to median value 
26 rows are removed 

SKEWNESS

*******************************************************AFTER*******************************************************

**** Box plot after removal of outliers****

PERCENTILE
*******************************************************AFTER*******************************************************

[' Inventory and accounts receivable/Net value',
 ' Current Liability to Current Assets',
 ' Cash/Current Liability',
 ' Cash Flow to Equity']


print_skewness_and_handle_outlier(7,10,bank_data_stage_1,'+',.95,True)

****Displaying first 5 rows of features that are having it skewness from 7 to 10****

**** Box plot before removal of outliers****

**** Displaying percentile of columns that are skewed from 7 to 10 ****

*******************************************************BEFORE*******************************************************

SKEWNESS

*******************************************************BEFORE*******************************************************


 **********Outliers beyond 0.95 percentile **********

2 minority class
2 rows of specific column  are imputed to median value 
219 rows are removed 

SKEWNESS

*******************************************************AFTER*******************************************************

**** Box plot after removal of outliers****

PERCENTILE
*******************************************************AFTER*******************************************************

[' Interest-bearing debt interest rate',
 ' Equity to Liability',
 ' Operating profit per person',
 ' Cash Flow Per Share',
 ' Net Worth Turnover Rate (times)',
 ' Total expense/Assets']


print_skewness_and_handle_outlier(4,7,bank_data_stage_1,'+',.8,True)

****Displaying first 5 rows of features that are having it skewness from 4 to 7****

**** Box plot before removal of outliers****

**** Displaying percentile of columns that are skewed from 4 to 7 ****

*******************************************************BEFORE*******************************************************

SKEWNESS

*******************************************************BEFORE*******************************************************


 **********Outliers beyond 0.8 percentile **********

0 minority class
0 rows of specific column  are imputed to median value 
1 rows are removed 

SKEWNESS

*******************************************************AFTER*******************************************************

**** Box plot after removal of outliers****

PERCENTILE
*******************************************************AFTER*******************************************************

[' Net Value Per Share (B)', ' Persistent EPS in the Last Four Seasons']


print_skewness_and_handle_outlier(-10,-2,bank_data_stage_1,'-',.01)

****Displaying first 5 rows of features that are having it skewness from -10 to -2****

**** Box plot before removal of outliers****

**** Displaying percentile of columns that are skewed from -10 to -2 ****

*******************************************************BEFORE*******************************************************

SKEWNESS

*******************************************************BEFORE*******************************************************


 **********Outliers beyond 0.01 percentile **********

0 minority class
0 rows of specific column  are imputed to median value 
0 rows are removed 

SKEWNESS

*******************************************************AFTER*******************************************************

**** Box plot after removal of outliers****

PERCENTILE
*******************************************************AFTER*******************************************************

[' Operating Gross Margin']


print_skewness_and_handle_outlier(-20,-10,bank_data_stage_1,'-',.01,True)

****Displaying first 5 rows of features that are having it skewness from -20 to -10****

**** Box plot before removal of outliers****

**** Displaying percentile of columns that are skewed from -20 to -10 ****

*******************************************************BEFORE*******************************************************

SKEWNESS

*******************************************************BEFORE*******************************************************


 **********Outliers beyond 0.01 percentile **********

0 minority class
0 rows of specific column  are imputed to median value 
3 rows are removed 

SKEWNESS

*******************************************************AFTER*******************************************************

**** Box plot after removal of outliers****

PERCENTILE
*******************************************************AFTER*******************************************************

[' Interest Expense Ratio',
 ' Interest Coverage Ratio (Interest expense to EBIT)',
 ' No-credit Interval',
 ' Retained Earnings to Total Assets']


# Not removing any outliers Just skewness
print_skewness_and_handle_outlier(-40,-20,bank_data_stage_1,'-',.01,False) 
# Final argument is False in the above function that means outlier removal will not happen hence table will be displayed for analysis

****Displaying first 5 rows of features that are having it skewness from -40 to -20****

**** Box plot before removal of outliers****

**** Displaying percentile of columns that are skewed from -40 to -20 ****

*******************************************************BEFORE*******************************************************

SKEWNESS

*******************************************************BEFORE*******************************************************


 **********Outliers beyond 0.01 percentile **********

2 minority class
2 rows of specific column  are imputed to median value 
0 rows are removed 

SKEWNESS

*******************************************************AFTER*******************************************************

**** Box plot after removal of outliers****

PERCENTILE
*******************************************************AFTER*******************************************************

[" Net Income to Stockholder's Equity",
 ' Working Capital/Equity',
 ' Working capitcal Turnover Rate',
 ' After-tax Net Profit Growth Rate']


print_skewness_and_handle_outlier(-85,-50,bank_data_stage_1,'-',.01,True)

****Displaying first 5 rows of features that are having it skewness from -85 to -50****

**** Box plot before removal of outliers****

**** Displaying percentile of columns that are skewed from -85 to -50 ****

*******************************************************BEFORE*******************************************************

SKEWNESS

*******************************************************BEFORE*******************************************************


 **********Outliers beyond 0.01 percentile **********

0 minority class
0 rows of specific column  are imputed to median value 
0 rows are removed 

SKEWNESS

*******************************************************AFTER*******************************************************

**** Box plot after removal of outliers****

PERCENTILE
*******************************************************AFTER*******************************************************

[' Operating Profit Growth Rate', ' Operating Profit Rate']


X=bank_data_stage_1.drop('Bankrupt?',axis=1)
y=bank_data_stage_1["Bankrupt?"]
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42,stratify=y)
bank_data_stage_2=X_train.copy()
bank_data_stage_2["Bankrupt?"]=y_train
# Checking the shape of the training and testing sets
X_train.shape,y_train.shape,X_test.shape,y_test.shape,bank_data_stage_2.shape

((5172, 73), (5172,), (1294, 73), (1294,), (5172, 74))


# checking the classes balance in the split dataset
y_test.value_counts(),y_train.value_counts()

(0    1250
 1      44
 Name: Bankrupt?, dtype: int64,
 0    4996
 1     176
 Name: Bankrupt?, dtype: int64)


from sklearn.feature_selection import f_classif,SelectKBest 
fs = SelectKBest(score_func=f_classif,k=70)
# Apply feature selection
fs.fit(bank_data_stage_2.drop("Bankrupt?",axis=1),bank_data_stage_2['Bankrupt?'])
feat_score=pd.DataFrame(data={'score':fs.scores_,'p_value':np.round(fs.pvalues_,4)},index=[x for x in bank_data_stage_1.columns if x not in ['Bankrupt?']]).sort_values("score",ascending=False)
feat_score=feat_score.reset_index().rename(columns={'index':'columns'})
# feat_score.rename(columns={'index':'columns'})
feat_score


plt.figure(figsize=(12,5),dpi=200)
sns.barplot(data=feat_score.head(50),y='score',x='columns')
plt.xticks(rotation=90)
plt.show()


def gather_outliers(data,cols,show_table=True,IQR_const=[1.5,1.7,1.8,1.9,2,2.1,2.2,2.5,2.9,3]):
    '''
    data:dataframe
    cols:features
    show:it shows table if true (Default True)
    IQR_const:IQR scale for outlier detection default value is [1.5,1.7,1.8,1.9,2,2.1,2.2,2.5,2.9,3]
    TYPE:
    data:dataframe
    cols:list
    show:bool
    IQR_const:list

    RETURNS:
    This function returns a table of outliers if handled according to the 
    scale of IQR and calculates the total data lost due to imputation of 
    outliers with set of indices removed from the dataframe due to outlier handling

    '''
    
    outlier_table=pd.DataFrame()
    # Initializing the dict object to store amount of outlier occurrence in the particular feature
    outlier_dict={}
    # Iterating over the IQR scales
    for IQR_constant in IQR_const:
        # Initializing the dict object Key related to the IQR scale to store amount of outlier occurrence in the particular feature
        outlier_dict[f'{IQR_constant} * IQR']={}
        # Initializing set to store the indices of the outliers indices for the IQR scale
        total_indices=set()
        # Iterating over the features
        for i in cols:
            
            #Initializing set to store outlier indices for the particular feature
            outlier_index_per_feat=set()
            # Calculating the first quantile
            first_quantile=data[i].quantile(.25)
            # Calculating the third quantile
            third_quantile=data[i].quantile(.75)
            # Calculating the IQR
            IQR=third_quantile-first_quantile
            # Calculating the lower bound
            lower_limit=first_quantile - IQR*IQR_constant
            # Calculating the upper bound
            upper_limit=third_quantile + IQR*IQR_constant
            # Filetering the dataframe based on the upper bound
            upper_outliers=data.loc[data[i] >  upper_limit,cols+['Bankrupt?']]
            # Adding the indices of the upper outliers to the set
            outlier_index_per_feat.update(upper_outliers.index)
            total_indices.update(upper_outliers.index)
            if show_table:
                display(upper_outliers)
            # Filetering the dataframe based on the lower bound
            lower_outlier=data.loc[data[i] <  lower_limit,cols+['Bankrupt?']]
            outlier_index_per_feat.update(lower_outlier.index)
            total_indices.update(lower_outlier.index)
            if show_table:
                display(lower_outlier)
            # Storing percentage of outlier occurrence of the particular feature
            outlier_dict[f'{IQR_constant} * IQR'][i]=np.round((len(outlier_index_per_feat)/data.shape[0])*100,2)
        # Storing the total amount of outlier occurrence for the particular IQR scale
        outlier_dict[f'{IQR_constant} * IQR']["Total"]=np.round((len(total_indices)/data.shape[0])*100,2)
    return pd.DataFrame(outlier_dict),total_indices


# Calculating all the outliers occerence  
outlier_indices=gather_outliers(bank_data_stage_2[bank_data_stage_2['Bankrupt?']==0],skewness_Range(-80,86),False)
# Sorting the first column of the outlier table ascendingly
outlier_indices[0].sort_values('1.5 * IQR',inplace=True)
ascendingly_order_outlier_occr_feat=outlier_indices[0].index
outlier_indices[0].style.bar()


outlier_indices[0].iloc[10:,].T.style.bar()


# Storing first 15 features that have less outlier occurrence
first_15_col=list(ascendingly_order_outlier_occr_feat[:15])
# Printing the table of outlier occurence of the first 15 features having low outlier occurrence 
gather_outliers(bank_data_stage_2[bank_data_stage_2['Bankrupt?']==0],first_15_col,False)[0].style.bar()


plt.figure(figsize=(12,10),dpi=200)
sns.lineplot(y=outlier_indices[0].loc["Total"].values,x=outlier_indices[0].columns)


graph_label_title("IQR scale to handle Outlier ",'Percentage of data lost','IQR Scale vs % of data Lost')

# outlier_indices[0].iloc[:,[1]].values


# CAUTION!!!! This function only imputes outliers of majority class
def impute_outliers(data,IQR_constant,cols):
    '''
    data:dataframe
    IQR_constant:IQR scale for outlier detection
    cols:features
    TYPE:
    data:dataframe
    IQR_constant:float
    cols:list

    Returns:
    This function returns the dataframe with imputed outliers 
    CAUTION!!!! it only imputes outliers of majority class
    '''
    # Separating dataframe according to the class (Minority class) or (Majority class)
    condition_0=data["Bankrupt?"]==0
    # Store the majority class 
    data_0=data[condition_0]
    # Store the minority class
    data_1=data[~ condition_0]
    # Initializing the set to store the indices of the outliers
    outlier_indices=set()
    # Iterating over the features
    for i in cols:
        # Calculating the first quantile
        first_quantile=data[i].quantile(.25)
        # Calculating the third quantile
        third_quantile=data[i].quantile(.75)
        # Calculating the IQR
        IQR=third_quantile-first_quantile
        # Calculating the lower bound
        lower_limit=first_quantile - IQR*IQR_constant
        # Calculating the upper bound
        upper_limit=third_quantile + IQR*IQR_constant
        # Filetering the dataframe based on the upper bound
        upper_outliers=data_0.loc[data_0[i] >  upper_limit,cols]
        # Adding the indices of the upper outliers to the set
        outlier_indices.update(upper_outliers.index)
        # Removing the outliers from dataframe based on the upper bound
        data_0.drop(upper_outliers.index,inplace=True)
        # Filetering the dataframe based on the lower bound
        lower_outlier=data_0.loc[data_0[i] <  lower_limit,cols]
        # Adding the indices of the lower outliers to the set
        outlier_indices.update(lower_outlier.index)
        # Removing the outliers from dataframe based on the lower bound
        data_0.drop(lower_outlier.index,inplace=True)
   
    return pd.concat([data_0,data_1])


# Storing the first 33 outliers having less outlier occurrence
first_33_col=list(ascendingly_order_outlier_occr_feat[:33])
gather_outliers(bank_data_stage_2[bank_data_stage_2['Bankrupt?']==0],first_33_col,False,IQR_const=[2.2,2.5,2.9,3,3.1,3.2,3.3,3.4,3.5,3.6,3.7,3.8,3.9,4.0,4.1])[0].style.bar()


def check_best_IQR_scale(model,sampling_method,IQR_range=[1.5,1.6,1.7,1.8,1.9,2.0,2.1,2.2,2.5]):
    '''
    model: estimator
    sampling_method:Over_sampling or Over sampling Object
    IQR_range: Range of IQR Scale
    
    This function Iterates through the IQR scales and check the model performance and stores the performance
    This function returns the model performance of all the IQR scales in DataFrame  Object
    '''
#     Dictionary to store the model evaluation metrics of IQR scales
    data_dict={}
#     For loop to iterate through out the IQR scales supplied 
    for i in IQR_range:
# Store the Data frame that  is filtered by outliers beyond the limit of IQR scale i.e for examples 1.5,1.6,1.7.. and so on
        dataframe=impute_outliers(bank_data_stage_2,i,list(ascendingly_order_outlier_occr_feat[:33]))
#     Stores  the model performance of data which outliers are filtered with the IQR scale
        data_dict[i]=get_model_performance(model,dataframe,sampling_method)
#    returning the collection of  model performance of all the range of IQR scales supplied
    return pd.DataFrame(data_dict).T
    
        
        
        
def get_model_performance(model,dataframe,sampling_method,display_report=False,custom_featrues=False,return_trained_model=False):
    '''
    
    model: estimator
    sampling_method:Over_sampling or Over sampling Object
    display_report: To display the classification report or not
    custom_featrues: To use the custom features or not
    
    TYPE:
    model: estimator object
    dataframe: DataFrame Object
    sampling_method: Over_sampling or Over sampling Object
    display_report: boolean
    custom_featrues: boolean or list of features

    1. This function takes the model and dataframe and Over sampling 
    Object and returns the model performance
    2. This function returns the model performance of the model in Dictionary Object
    '''
    # If custom features are not supplied then use the default features

    X_train=dataframe.drop("Bankrupt?",axis=1)
    y_train=dataframe["Bankrupt?"]
    
    # If custom features are supplied then model uses the selected features
    if custom_featrues != False:
        X_test_=X_test[custom_featrues]
        X_train=X_train[custom_featrues]
    else:
        X_test_=X_test
        
  # Scalliing the data
    scaller=StandardScaler()
    X_train_=scaller.fit_transform(X_train)
    X_train_,y_train_=sampling_method.fit_resample(X_train_,y_train)
    X_test_=scaller.transform(X_test_)

#    Fitting the model
    model.fit(X_train_,y_train_)
    
#    Predicting the model
    predictions=model.predict(X_test_)
    
#    Calculating the model performance
    f1=f1_score(y_test,predictions)
    pre=precision_score(y_test,predictions)
    reca=recall_score(y_test,predictions)
    acc=accuracy_score(y_test,predictions)
    
#   Displaying the metrics
    if display_report:
        print(f1,'f1score')
        print(acc,'accuracy')
        print(pre,'precision')
        print(reca,'recall')
        print(classification_report(y_test,predictions,))
        print(confusion_matrix(y_test,predictions,))
    if return_trained_model:
        return {'f1_score':f1,'precision':pre,'recall':reca,'accuracy':acc},model,X_test_
    else :
        return {'f1_score':f1,'precision':pre,'recall':reca,'accuracy':acc}
    
def print_metrics(predictions):
    '''
    predictions: predicted Data
    TYPE:
    predictions:  numpy array
    This function takes the predicted data and prints the metrics
    '''
    # Calculating the metrics
    f1=f1_score(y_test,predictions)
    pre=precision_score(y_test,predictions)
    reca=recall_score(y_test,predictions)
    acc=accuracy_score(y_test,predictions)
    
    # Displaying the metrics
    print(f1,'f1score')
    print(acc,'accuracy')
    print(pre,'precision')
    print(reca,'recall')
    print(classification_report(y_test,predictions,))
    print(confusion_matrix(y_test,predictions))
    return {'f1_score':f1,'precision':pre,'recall':reca,'accuracy':acc}

def plot_best_score(data,model_name):
    print(data["f1_score"].max())
    max_score=data[data["f1_score"]==data["f1_score"].max()]
    display(data[data["f1_score"]==data["f1_score"].max()])
    precision=max_score["precision"].values
    recall=max_score["recall"].values
    f1_score=max_score["f1_score"].values
    max_index=max_score.index[0]


    plt.figure(figsize=(10,5),dpi=200) # for better resolution

    # plotting the metrics of model to visualize the performance
    plt.plot(data.index,data["f1_score"],'y',label="fi_score", linewidth = '5')
    plt.plot(data.index,data["recall"],'r',label="recall", linewidth = '5')
    plt.plot(data.index,data["precision"],'g',label="precision", linewidth = '5')
    
    # Highlighting the best part of curve
    plt.axvline(x=max_index,label="Max IQR scale",color ="orange")
    plt.axhline(y=f1_score,label="f1-score",color ="green")
    plt.axhline(y=precision,label="precision",color='red')
    plt.axhline(y=recall,label="recall",color='yellow')
    graph_label_title('IQR Scale value',"scale for metrics ", model_name)
    plt.legend()
    plt.show()

# Displaying the model performance metrics of IQR scale range between 2.1 to 3.5
    display(data.style.highlight_max())


# Tuning the model based on IQR scale and checking the best performance
chart_IQR_scale_tune_XGB=check_best_IQR_scale(XGBClassifier(),ADASYN(random_state=42,n_neighbors=5,n_jobs=-1),[2.0,2.2,2.3,2.5,3.3,3.4,3.7,3.8,3.9])

[12:37:16] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[12:37:23] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[12:37:30] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[12:37:38] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[12:37:45] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[12:37:53] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[12:38:02] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[12:38:12] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[12:38:21] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.


plot_best_score(chart_IQR_scale_tune_XGB,"Tuning by IQR scale for XGboost Model")

0.5299145299145299


# Storing the data treated with Outlier of IQR scale 3.8
data_xgb=impute_outliers(bank_data_stage_2,3.8,list(ascendingly_order_outlier_occr_feat[:33]))

# Calculating the overall performance of the model
get_model_performance(XGBClassifier(),
                      data_xgb,ADASYN(random_state=42,n_neighbors=5,n_jobs=-1),
                      display_report=True)

[10:48:27] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
0.5299145299145299 f1score
0.9574961360123647 accuracy
0.4246575342465753 precision
0.7045454545454546 recall
              precision    recall  f1-score   support

           0       0.99      0.97      0.98      1250
           1       0.42      0.70      0.53        44

    accuracy                           0.96      1294
   macro avg       0.71      0.84      0.75      1294
weighted avg       0.97      0.96      0.96      1294

[[1208   42]
 [  13   31]]

{'f1_score': 0.5299145299145299,
 'precision': 0.4246575342465753,
 'recall': 0.7045454545454546,
 'accuracy': 0.9574961360123647}


del f1_score


# Storing the data handled by outliers beyond IQR scale 3.8
data_xgb=impute_outliers(bank_data_stage_2,3.8,list(ascendingly_order_outlier_occr_feat[:33]))

# Storing the model performance metrics 
feature_selection_tune_xgb={}

# Starting feature selection from 49th ranked feature to last ranked feature
features_=list(feat_score['columns'].values[:49])

# Iterating from 49th ranked feature to 55th ranked feature
for i in feat_score['columns'].values[49:55]:
# Adding the next ranked feature to the list of features
    features_.append(i)
    
# Storing the model performance metrics for the current features set
    feature_selection_tune_xgb[len(features_)]=get_model_performance(XGBClassifier(),
                                                                     data_xgb,ADASYN(random_state=42,
                                                                                     n_neighbors=5
                                                                                     ,n_jobs=-1),
                                                                     display_report=False,
                                                                     custom_featrues=features_)

[14:19:25] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[14:19:31] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[14:19:35] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[14:19:41] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[14:19:46] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[14:19:51] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.


# Storing data stored in dictionary data in the form of dataframe 
feature_selection_tune_xgb_=pd.DataFrame(feature_selection_tune_xgb).T
plot_best_score(feature_selection_tune_xgb_,"Tuning by Feature Selection for XGboost Model")

0.5454545454545454


# Storing the  features selected from rank 1 to rank 52
features_for_XgBoost=list(feat_score['columns'].values[:52])

# Storing the data treated with Outlier of IQR scale 3.8 and features selected from rank 1 to rank 52
data_xgb=impute_outliers(bank_data_stage_2,3.8,list(ascendingly_order_outlier_occr_feat[:33]))

# Checking the overall performance of the model
get_model_performance(XGBClassifier(),
                      data_xgb,ADASYN(random_state=42,n_neighbors=5,n_jobs=-1),
                      display_report=True,
                      custom_featrues= features_for_XgBoost)

[15:50:51] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
0.5454545454545454 f1score
0.9574961360123647 accuracy
0.42857142857142855 precision
0.75 recall
              precision    recall  f1-score   support

           0       0.99      0.96      0.98      1250
           1       0.43      0.75      0.55        44

    accuracy                           0.96      1294
   macro avg       0.71      0.86      0.76      1294
weighted avg       0.97      0.96      0.96      1294

[[1206   44]
 [  11   33]]

{'f1_score': 0.5454545454545454,
 'precision': 0.42857142857142855,
 'recall': 0.75,
 'accuracy': 0.9574961360123647}


# Storing the data suitable for Xgboost model giving best performance
X_train=data_xgb[features_for_XgBoost]
y_train=data_xgb["Bankrupt?"]

# Scaling the data
scaller=StandardScaler()
X_train_=scaller.fit_transform(X_train)
X_test_=scaller.transform(X_test[features_for_XgBoost])

# OverSampling the data to balance the class
X_train_,y_train_=ADASYN(random_state=42,n_neighbors=5,n_jobs=-1).fit_resample(X_train_,y_train)

# Storing the parameters for Xgboost model for tuning
param_grid={'n_estimators':[100,200,300],'learning_rate':[0.2,0.3,0.4,0.5,0.6],'max_depth':[2,3,4,5,6],}

model=XGBClassifier()

# Creating an instance of the GridSearchCV with the parameters and the model
grid=RandomizedSearchCV(estimator=model,param_distributions= param_grid
                        ,cv=5,verbose=1,
                        scoring='f1')

# Fitting the model
grid.fit(X_train_,y_train_)
print('\n Best Parameters\n',grid.best_params_)
# Storing the best parameters
model=grid.best_estimator_

# Printing the model evaluation metrics
print_metrics(predictions=model.predict(X_test_))

Fitting 5 folds for each of 10 candidates, totalling 50 fits
[16:44:26] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:44:30] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:44:33] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:44:37] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:44:40] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:44:44] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:44:52] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:44:58] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:45:05] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:45:12] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:45:19] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:45:28] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:45:37] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:45:46] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:45:55] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:46:03] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:46:10] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:46:15] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:46:19] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:46:27] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:46:32] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:46:42] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:46:51] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:47:02] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:47:13] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:47:21] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:47:23] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:47:25] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:47:27] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:47:29] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:47:30] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:47:35] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:47:39] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:47:44] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:47:49] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:47:54] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:47:57] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:48:00] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:48:03] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:48:06] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:48:09] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:48:12] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:48:15] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:48:18] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:48:21] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:48:24] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:48:26] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:48:27] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:48:29] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:48:31] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
[16:48:33] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.

 Best Parameters
 {'n_estimators': 300, 'max_depth': 5, 'learning_rate': 0.2}
0.5299145299145299 f1score
0.9574961360123647 accuracy
0.4246575342465753 precision
0.7045454545454546 recall
              precision    recall  f1-score   support

           0       0.99      0.97      0.98      1250
           1       0.42      0.70      0.53        44

    accuracy                           0.96      1294
   macro avg       0.71      0.84      0.75      1294
weighted avg       0.97      0.96      0.96      1294

[[1208   42]
 [  13   31]]

{'f1_score': 0.5299145299145299,
 'precision': 0.4246575342465753,
 'recall': 0.7045454545454546,
 'accuracy': 0.9574961360123647}


# Storing the perforamces of model of all values in range of IQR values from range 2.1 to 3.5
chart_IQR_scale_tune_RF=check_best_IQR_scale(RandomForestClassifier(),
                                             ADASYN(random_state=42,n_neighbors=5,n_jobs=-1),
                                             [2.1,2.2,2.3,2.4,2.5,2.6,2.7,2.8,2.9,3.0,])


def plot_best_score(data,model_name):
    print(data["f1_score"].max())
    max_score=data[data["f1_score"]==data["f1_score"].max()]
    display(data[data["f1_score"]==data["f1_score"].max()])
    precision=max_score["precision"].values
    recall=max_score["recall"].values
    f1_score=max_score["f1_score"].values
    max_index=max_score.index[0]


    plt.figure(figsize=(10,5),dpi=200) # for better resolution

    # plotting the metrics of model to visualize the performance
    plt.plot(data.index,data["f1_score"],'y',label="fi_score", linewidth = '5')
    plt.plot(data.index,data["recall"],'r',label="recall", linewidth = '5')
    plt.plot(data.index,data["precision"],'g',label="precision", linewidth = '5')
    
    # Highlighting the best part of curve
    plt.axvline(x=max_index,label="Max IQR scale",color ="orange")
    plt.axhline(y=f1_score,label="f1-score",color ="green")
    plt.axhline(y=precision,label="precision",color='red')
    plt.axhline(y=recall,label="recall",color='yellow')
    graph_label_title('IQR Scale value',"scale for metrics ", model_name)
    plt.legend()
    plt.show()

# Displaying the model performance metrics of IQR scale range between 2.1 to 3.5
    display(data.style.highlight_max())


plot_best_score(chart_IQR_scale_tune_RF,"Tuning by IQR scale for Random Forest model")

0.46357615894039733


# Storing best IQR scale value that gives best performance for Random Forest model
IQR_scale=2.4

# Storing the  data treated with Outlier of IQR scale 2.5 
data_rf=impute_outliers(bank_data_stage_2,IQR_scale,list(ascendingly_order_outlier_occr_feat[:33]))

# Checking the overall performance of the model with best IQR scale
get_model_performance(RandomForestClassifier(),
                      data_rf,ADASYN(random_state=42,n_neighbors=5,n_jobs=-1)
                      ,display_report=True)

0.44295302013422816 f1score
0.9358578052550232 accuracy
0.3142857142857143 precision
0.75 recall
              precision    recall  f1-score   support

           0       0.99      0.94      0.97      1250
           1       0.31      0.75      0.44        44

    accuracy                           0.94      1294
   macro avg       0.65      0.85      0.70      1294
weighted avg       0.97      0.94      0.95      1294

[[1178   72]
 [  11   33]]

{'f1_score': 0.44295302013422816,
 'precision': 0.3142857142857143,
 'recall': 0.75,
 'accuracy': 0.9358578052550232}


# Storing the data handled by outliers beyond IQR scale 2.5
feature_selection_tune_rf={}
# Storing first 60 ranked features
features_=list(feat_score['columns'].values[:60])
# Iterating through the features from rank 55 to last ranked features
for i in feat_score['columns'].values[60:]:
    # Adding next ranked feature to the feature  list
    features_.append(i)
    # Storing the model performance metrics for the features
    feature_selection_tune_rf[len(features_)]=get_model_performance(RandomForestClassifier()
                                                                    ,data_rf,ADASYN(
                                                                        random_state=42,
                                                                        n_neighbors=5,
                                                                        n_jobs=-1),
                                                                    display_report=False,
                                                                    custom_featrues=features_)


feature_selection_tune_rf_=pd.DataFrame(feature_selection_tune_rf).T
plot_best_score(feature_selection_tune_rf_,"Tuning by Feature Sleection for Random Forest model")

0.45333333333333337


features_for_Rf=list(feat_score['columns'].values[:69])
get_model_performance(RandomForestClassifier(),
                      data_rf,ADASYN(random_state=42,n_neighbors=5,n_jobs=-1)
                      ,display_report=True,custom_featrues=features_for_Rf)

0.45070422535211263 f1score
0.9397217928902627 accuracy
0.32653061224489793 precision
0.7272727272727273 recall
              precision    recall  f1-score   support

           0       0.99      0.95      0.97      1250
           1       0.33      0.73      0.45        44

    accuracy                           0.94      1294
   macro avg       0.66      0.84      0.71      1294
weighted avg       0.97      0.94      0.95      1294

[[1184   66]
 [  12   32]]

{'f1_score': 0.45070422535211263,
 'precision': 0.32653061224489793,
 'recall': 0.7272727272727273,
 'accuracy': 0.9397217928902627}


# Storing Xtrain set and ytrain set for Random Forest model
X_train=data_rf.drop("Bankrupt?",axis=1)
y_train=data_rf["Bankrupt?"]

# Scalining the data
scaller=StandardScaler()
X_train_=scaller.fit_transform(X_train)
X_test_=scaller.transform(X_test)

# OverSampling the data
X_train_,y_train_=ADASYN(random_state=42,n_neighbors=5,n_jobs=-1).fit_resample(X_train_,y_train)

# Storing the best parameters for tuning Random Forest model
param_grid={'bootstrap': [True, False],
 'max_depth': [10, 20, 30, 40, 50, 60, 70, 80, 90, 100, None],
 'max_features': ['auto', 'sqrt'],
 'min_samples_leaf': [1, 2, 4],
 'min_samples_split': [2, 5, 10],
 'n_estimators': [30,40,50,70,80,100,1400, 1600, 1800, 2000]}

# Initializing the Random Forest model
model=RandomForestClassifier()

# Initializing the Randomised Search CV object
grid=RandomizedSearchCV(estimator=model,param_distributions= param_grid
                        ,cv=5,verbose=10,
                        scoring='neg_mean_absolute_error')

# Fitting the model
grid.fit(X_train_,y_train_)
print(grid.best_params_)

# Storing the best parameters for Random Forest model
model=grid.best_estimator_

# Printing the best parameters for Random Forest model
print_metrics(predictions=model.predict(X_test_))

Fitting 5 folds for each of 10 candidates, totalling 50 fits
[CV 1/5; 1/10] START bootstrap=False, max_depth=60, max_features=sqrt, min_samples_leaf=4, min_samples_split=2, n_estimators=40
[CV 1/5; 1/10] END bootstrap=False, max_depth=60, max_features=sqrt, min_samples_leaf=4, min_samples_split=2, n_estimators=40;, score=-0.061 total time=   1.7s
[CV 2/5; 1/10] START bootstrap=False, max_depth=60, max_features=sqrt, min_samples_leaf=4, min_samples_split=2, n_estimators=40
[CV 2/5; 1/10] END bootstrap=False, max_depth=60, max_features=sqrt, min_samples_leaf=4, min_samples_split=2, n_estimators=40;, score=-0.044 total time=   1.7s
[CV 3/5; 1/10] START bootstrap=False, max_depth=60, max_features=sqrt, min_samples_leaf=4, min_samples_split=2, n_estimators=40
[CV 3/5; 1/10] END bootstrap=False, max_depth=60, max_features=sqrt, min_samples_leaf=4, min_samples_split=2, n_estimators=40;, score=-0.051 total time=   2.2s
[CV 4/5; 1/10] START bootstrap=False, max_depth=60, max_features=sqrt, min_samples_leaf=4, min_samples_split=2, n_estimators=40
[CV 4/5; 1/10] END bootstrap=False, max_depth=60, max_features=sqrt, min_samples_leaf=4, min_samples_split=2, n_estimators=40;, score=-0.098 total time=   1.8s
[CV 5/5; 1/10] START bootstrap=False, max_depth=60, max_features=sqrt, min_samples_leaf=4, min_samples_split=2, n_estimators=40
[CV 5/5; 1/10] END bootstrap=False, max_depth=60, max_features=sqrt, min_samples_leaf=4, min_samples_split=2, n_estimators=40;, score=-0.038 total time=   2.6s
[CV 1/5; 2/10] START bootstrap=True, max_depth=20, max_features=sqrt, min_samples_leaf=1, min_samples_split=10, n_estimators=40
[CV 1/5; 2/10] END bootstrap=True, max_depth=20, max_features=sqrt, min_samples_leaf=1, min_samples_split=10, n_estimators=40;, score=-0.054 total time=   1.8s
[CV 2/5; 2/10] START bootstrap=True, max_depth=20, max_features=sqrt, min_samples_leaf=1, min_samples_split=10, n_estimators=40
[CV 2/5; 2/10] END bootstrap=True, max_depth=20, max_features=sqrt, min_samples_leaf=1, min_samples_split=10, n_estimators=40;, score=-0.051 total time=   1.3s
[CV 3/5; 2/10] START bootstrap=True, max_depth=20, max_features=sqrt, min_samples_leaf=1, min_samples_split=10, n_estimators=40
[CV 3/5; 2/10] END bootstrap=True, max_depth=20, max_features=sqrt, min_samples_leaf=1, min_samples_split=10, n_estimators=40;, score=-0.052 total time=   1.1s
[CV 4/5; 2/10] START bootstrap=True, max_depth=20, max_features=sqrt, min_samples_leaf=1, min_samples_split=10, n_estimators=40
[CV 4/5; 2/10] END bootstrap=True, max_depth=20, max_features=sqrt, min_samples_leaf=1, min_samples_split=10, n_estimators=40;, score=-0.100 total time=   1.2s
[CV 5/5; 2/10] START bootstrap=True, max_depth=20, max_features=sqrt, min_samples_leaf=1, min_samples_split=10, n_estimators=40
[CV 5/5; 2/10] END bootstrap=True, max_depth=20, max_features=sqrt, min_samples_leaf=1, min_samples_split=10, n_estimators=40;, score=-0.042 total time=   1.1s
[CV 1/5; 3/10] START bootstrap=True, max_depth=60, max_features=auto, min_samples_leaf=1, min_samples_split=2, n_estimators=70
[CV 1/5; 3/10] END bootstrap=True, max_depth=60, max_features=auto, min_samples_leaf=1, min_samples_split=2, n_estimators=70;, score=-0.063 total time=   2.1s
[CV 2/5; 3/10] START bootstrap=True, max_depth=60, max_features=auto, min_samples_leaf=1, min_samples_split=2, n_estimators=70
[CV 2/5; 3/10] END bootstrap=True, max_depth=60, max_features=auto, min_samples_leaf=1, min_samples_split=2, n_estimators=70;, score=-0.043 total time=   2.8s
[CV 3/5; 3/10] START bootstrap=True, max_depth=60, max_features=auto, min_samples_leaf=1, min_samples_split=2, n_estimators=70
[CV 3/5; 3/10] END bootstrap=True, max_depth=60, max_features=auto, min_samples_leaf=1, min_samples_split=2, n_estimators=70;, score=-0.049 total time=   2.5s
[CV 4/5; 3/10] START bootstrap=True, max_depth=60, max_features=auto, min_samples_leaf=1, min_samples_split=2, n_estimators=70
[CV 4/5; 3/10] END bootstrap=True, max_depth=60, max_features=auto, min_samples_leaf=1, min_samples_split=2, n_estimators=70;, score=-0.086 total time=   2.7s
[CV 5/5; 3/10] START bootstrap=True, max_depth=60, max_features=auto, min_samples_leaf=1, min_samples_split=2, n_estimators=70
[CV 5/5; 3/10] END bootstrap=True, max_depth=60, max_features=auto, min_samples_leaf=1, min_samples_split=2, n_estimators=70;, score=-0.039 total time=   2.2s
[CV 1/5; 4/10] START bootstrap=True, max_depth=40, max_features=auto, min_samples_leaf=2, min_samples_split=2, n_estimators=40
[CV 1/5; 4/10] END bootstrap=True, max_depth=40, max_features=auto, min_samples_leaf=2, min_samples_split=2, n_estimators=40;, score=-0.065 total time=   1.1s
[CV 2/5; 4/10] START bootstrap=True, max_depth=40, max_features=auto, min_samples_leaf=2, min_samples_split=2, n_estimators=40
[CV 2/5; 4/10] END bootstrap=True, max_depth=40, max_features=auto, min_samples_leaf=2, min_samples_split=2, n_estimators=40;, score=-0.047 total time=   1.2s
[CV 3/5; 4/10] START bootstrap=True, max_depth=40, max_features=auto, min_samples_leaf=2, min_samples_split=2, n_estimators=40
[CV 3/5; 4/10] END bootstrap=True, max_depth=40, max_features=auto, min_samples_leaf=2, min_samples_split=2, n_estimators=40;, score=-0.052 total time=   1.4s
[CV 4/5; 4/10] START bootstrap=True, max_depth=40, max_features=auto, min_samples_leaf=2, min_samples_split=2, n_estimators=40
[CV 4/5; 4/10] END bootstrap=True, max_depth=40, max_features=auto, min_samples_leaf=2, min_samples_split=2, n_estimators=40;, score=-0.096 total time=   1.5s
[CV 5/5; 4/10] START bootstrap=True, max_depth=40, max_features=auto, min_samples_leaf=2, min_samples_split=2, n_estimators=40
[CV 5/5; 4/10] END bootstrap=True, max_depth=40, max_features=auto, min_samples_leaf=2, min_samples_split=2, n_estimators=40;, score=-0.036 total time=   2.7s
[CV 1/5; 5/10] START bootstrap=False, max_depth=None, max_features=sqrt, min_samples_leaf=4, min_samples_split=5, n_estimators=30
[CV 1/5; 5/10] END bootstrap=False, max_depth=None, max_features=sqrt, min_samples_leaf=4, min_samples_split=5, n_estimators=30;, score=-0.058 total time=   3.2s
[CV 2/5; 5/10] START bootstrap=False, max_depth=None, max_features=sqrt, min_samples_leaf=4, min_samples_split=5, n_estimators=30
[CV 2/5; 5/10] END bootstrap=False, max_depth=None, max_features=sqrt, min_samples_leaf=4, min_samples_split=5, n_estimators=30;, score=-0.048 total time=   1.9s
[CV 3/5; 5/10] START bootstrap=False, max_depth=None, max_features=sqrt, min_samples_leaf=4, min_samples_split=5, n_estimators=30
[CV 3/5; 5/10] END bootstrap=False, max_depth=None, max_features=sqrt, min_samples_leaf=4, min_samples_split=5, n_estimators=30;, score=-0.051 total time=   2.5s
[CV 4/5; 5/10] START bootstrap=False, max_depth=None, max_features=sqrt, min_samples_leaf=4, min_samples_split=5, n_estimators=30
[CV 4/5; 5/10] END bootstrap=False, max_depth=None, max_features=sqrt, min_samples_leaf=4, min_samples_split=5, n_estimators=30;, score=-0.084 total time=   1.6s
[CV 5/5; 5/10] START bootstrap=False, max_depth=None, max_features=sqrt, min_samples_leaf=4, min_samples_split=5, n_estimators=30
[CV 5/5; 5/10] END bootstrap=False, max_depth=None, max_features=sqrt, min_samples_leaf=4, min_samples_split=5, n_estimators=30;, score=-0.038 total time=   1.3s
[CV 1/5; 6/10] START bootstrap=True, max_depth=60, max_features=auto, min_samples_leaf=4, min_samples_split=2, n_estimators=30
[CV 1/5; 6/10] END bootstrap=True, max_depth=60, max_features=auto, min_samples_leaf=4, min_samples_split=2, n_estimators=30;, score=-0.057 total time=   1.3s
[CV 2/5; 6/10] START bootstrap=True, max_depth=60, max_features=auto, min_samples_leaf=4, min_samples_split=2, n_estimators=30
[CV 2/5; 6/10] END bootstrap=True, max_depth=60, max_features=auto, min_samples_leaf=4, min_samples_split=2, n_estimators=30;, score=-0.049 total time=   0.7s
[CV 3/5; 6/10] START bootstrap=True, max_depth=60, max_features=auto, min_samples_leaf=4, min_samples_split=2, n_estimators=30
[CV 3/5; 6/10] END bootstrap=True, max_depth=60, max_features=auto, min_samples_leaf=4, min_samples_split=2, n_estimators=30;, score=-0.056 total time=   0.9s
[CV 4/5; 6/10] START bootstrap=True, max_depth=60, max_features=auto, min_samples_leaf=4, min_samples_split=2, n_estimators=30
[CV 4/5; 6/10] END bootstrap=True, max_depth=60, max_features=auto, min_samples_leaf=4, min_samples_split=2, n_estimators=30;, score=-0.093 total time=   0.8s
[CV 5/5; 6/10] START bootstrap=True, max_depth=60, max_features=auto, min_samples_leaf=4, min_samples_split=2, n_estimators=30
[CV 5/5; 6/10] END bootstrap=True, max_depth=60, max_features=auto, min_samples_leaf=4, min_samples_split=2, n_estimators=30;, score=-0.044 total time=   0.9s
[CV 1/5; 7/10] START bootstrap=True, max_depth=40, max_features=sqrt, min_samples_leaf=1, min_samples_split=10, n_estimators=30
[CV 1/5; 7/10] END bootstrap=True, max_depth=40, max_features=sqrt, min_samples_leaf=1, min_samples_split=10, n_estimators=30;, score=-0.062 total time=   1.1s
[CV 2/5; 7/10] START bootstrap=True, max_depth=40, max_features=sqrt, min_samples_leaf=1, min_samples_split=10, n_estimators=30
[CV 2/5; 7/10] END bootstrap=True, max_depth=40, max_features=sqrt, min_samples_leaf=1, min_samples_split=10, n_estimators=30;, score=-0.049 total time=   1.2s
[CV 3/5; 7/10] START bootstrap=True, max_depth=40, max_features=sqrt, min_samples_leaf=1, min_samples_split=10, n_estimators=30
[CV 3/5; 7/10] END bootstrap=True, max_depth=40, max_features=sqrt, min_samples_leaf=1, min_samples_split=10, n_estimators=30;, score=-0.050 total time=   0.8s
[CV 4/5; 7/10] START bootstrap=True, max_depth=40, max_features=sqrt, min_samples_leaf=1, min_samples_split=10, n_estimators=30
[CV 4/5; 7/10] END bootstrap=True, max_depth=40, max_features=sqrt, min_samples_leaf=1, min_samples_split=10, n_estimators=30;, score=-0.093 total time=   0.9s
[CV 5/5; 7/10] START bootstrap=True, max_depth=40, max_features=sqrt, min_samples_leaf=1, min_samples_split=10, n_estimators=30
[CV 5/5; 7/10] END bootstrap=True, max_depth=40, max_features=sqrt, min_samples_leaf=1, min_samples_split=10, n_estimators=30;, score=-0.042 total time=   0.9s
[CV 1/5; 8/10] START bootstrap=False, max_depth=40, max_features=auto, min_samples_leaf=1, min_samples_split=5, n_estimators=1600
[CV 1/5; 8/10] END bootstrap=False, max_depth=40, max_features=auto, min_samples_leaf=1, min_samples_split=5, n_estimators=1600;, score=-0.058 total time= 1.5min
[CV 2/5; 8/10] START bootstrap=False, max_depth=40, max_features=auto, min_samples_leaf=1, min_samples_split=5, n_estimators=1600
[CV 2/5; 8/10] END bootstrap=False, max_depth=40, max_features=auto, min_samples_leaf=1, min_samples_split=5, n_estimators=1600;, score=-0.037 total time= 1.6min
[CV 3/5; 8/10] START bootstrap=False, max_depth=40, max_features=auto, min_samples_leaf=1, min_samples_split=5, n_estimators=1600
[CV 3/5; 8/10] END bootstrap=False, max_depth=40, max_features=auto, min_samples_leaf=1, min_samples_split=5, n_estimators=1600;, score=-0.042 total time= 1.4min
[CV 4/5; 8/10] START bootstrap=False, max_depth=40, max_features=auto, min_samples_leaf=1, min_samples_split=5, n_estimators=1600
[CV 4/5; 8/10] END bootstrap=False, max_depth=40, max_features=auto, min_samples_leaf=1, min_samples_split=5, n_estimators=1600;, score=-0.088 total time= 1.4min
[CV 5/5; 8/10] START bootstrap=False, max_depth=40, max_features=auto, min_samples_leaf=1, min_samples_split=5, n_estimators=1600
[CV 5/5; 8/10] END bootstrap=False, max_depth=40, max_features=auto, min_samples_leaf=1, min_samples_split=5, n_estimators=1600;, score=-0.032 total time= 1.3min
[CV 1/5; 9/10] START bootstrap=True, max_depth=70, max_features=auto, min_samples_leaf=2, min_samples_split=5, n_estimators=100
[CV 1/5; 9/10] END bootstrap=True, max_depth=70, max_features=auto, min_samples_leaf=2, min_samples_split=5, n_estimators=100;, score=-0.058 total time=   2.9s
[CV 2/5; 9/10] START bootstrap=True, max_depth=70, max_features=auto, min_samples_leaf=2, min_samples_split=5, n_estimators=100
[CV 2/5; 9/10] END bootstrap=True, max_depth=70, max_features=auto, min_samples_leaf=2, min_samples_split=5, n_estimators=100;, score=-0.047 total time=   2.8s
[CV 3/5; 9/10] START bootstrap=True, max_depth=70, max_features=auto, min_samples_leaf=2, min_samples_split=5, n_estimators=100
[CV 3/5; 9/10] END bootstrap=True, max_depth=70, max_features=auto, min_samples_leaf=2, min_samples_split=5, n_estimators=100;, score=-0.050 total time=   2.8s
[CV 4/5; 9/10] START bootstrap=True, max_depth=70, max_features=auto, min_samples_leaf=2, min_samples_split=5, n_estimators=100
[CV 4/5; 9/10] END bootstrap=True, max_depth=70, max_features=auto, min_samples_leaf=2, min_samples_split=5, n_estimators=100;, score=-0.087 total time=   2.7s
[CV 5/5; 9/10] START bootstrap=True, max_depth=70, max_features=auto, min_samples_leaf=2, min_samples_split=5, n_estimators=100
[CV 5/5; 9/10] END bootstrap=True, max_depth=70, max_features=auto, min_samples_leaf=2, min_samples_split=5, n_estimators=100;, score=-0.037 total time=   2.9s
[CV 1/5; 10/10] START bootstrap=True, max_depth=20, max_features=sqrt, min_samples_leaf=4, min_samples_split=5, n_estimators=1800
[CV 1/5; 10/10] END bootstrap=True, max_depth=20, max_features=sqrt, min_samples_leaf=4, min_samples_split=5, n_estimators=1800;, score=-0.062 total time=  53.0s
[CV 2/5; 10/10] START bootstrap=True, max_depth=20, max_features=sqrt, min_samples_leaf=4, min_samples_split=5, n_estimators=1800
[CV 2/5; 10/10] END bootstrap=True, max_depth=20, max_features=sqrt, min_samples_leaf=4, min_samples_split=5, n_estimators=1800;, score=-0.048 total time=  50.5s
[CV 3/5; 10/10] START bootstrap=True, max_depth=20, max_features=sqrt, min_samples_leaf=4, min_samples_split=5, n_estimators=1800
[CV 3/5; 10/10] END bootstrap=True, max_depth=20, max_features=sqrt, min_samples_leaf=4, min_samples_split=5, n_estimators=1800;, score=-0.054 total time=  51.2s
[CV 4/5; 10/10] START bootstrap=True, max_depth=20, max_features=sqrt, min_samples_leaf=4, min_samples_split=5, n_estimators=1800
[CV 4/5; 10/10] END bootstrap=True, max_depth=20, max_features=sqrt, min_samples_leaf=4, min_samples_split=5, n_estimators=1800;, score=-0.094 total time=  50.6s
[CV 5/5; 10/10] START bootstrap=True, max_depth=20, max_features=sqrt, min_samples_leaf=4, min_samples_split=5, n_estimators=1800
[CV 5/5; 10/10] END bootstrap=True, max_depth=20, max_features=sqrt, min_samples_leaf=4, min_samples_split=5, n_estimators=1800;, score=-0.035 total time=  53.2s
{'n_estimators': 1600, 'min_samples_split': 5, 'min_samples_leaf': 1, 'max_features': 'auto', 'max_depth': 40, 'bootstrap': False}
0.4477611940298507 f1score
0.9428129829984544 accuracy
0.3333333333333333 precision
0.6818181818181818 recall
              precision    recall  f1-score   support

           0       0.99      0.95      0.97      1250
           1       0.33      0.68      0.45        44

    accuracy                           0.94      1294
   macro avg       0.66      0.82      0.71      1294
weighted avg       0.97      0.94      0.95      1294

[[1190   60]
 [  14   30]]

{'f1_score': 0.4477611940298507,
 'precision': 0.3333333333333333,
 'recall': 0.6818181818181818,
 'accuracy': 0.9428129829984544}


# Storing the model performance for Random Forest model for IQR range of 1.5 to 3.9
chart_IQR_scale_tune_GDBoost=check_best_IQR_scale(GradientBoostingClassifier()
                                                  ,ADASYN(random_state=42,
                                                          n_neighbors=5,
                                                          n_jobs=-1),
                                                  [2.3,2.4,2.5,2.6,2.8
                                                   ,3.0,3.1,3.2,3.3,3.4,3.5,3.6,3.7,3.8,3.9,4,4.1,4.2])


plot_best_score(chart_IQR_scale_tune_GDBoost,"Tuning by IQR scale for Gradient Boost model")

0.4


# Storing Optimum value of IQR scale for Gradient Boost model
IQR_scale=3.6

# Storing the data after removing outlier beyond IQR scale i.e 3.7
data_GBoost=impute_outliers(bank_data_stage_2,IQR_scale,list(ascendingly_order_outlier_occr_feat[:33]))

# 
get_model_performance(GradientBoostingClassifier(),
                      data_GBoost,ADASYN(random_state=42,n_neighbors=5,n_jobs=-1)
                      ,display_report=True)

0.4 f1score
0.9142194744976816 accuracy
0.2624113475177305 precision
0.8409090909090909 recall
              precision    recall  f1-score   support

           0       0.99      0.92      0.95      1250
           1       0.26      0.84      0.40        44

    accuracy                           0.91      1294
   macro avg       0.63      0.88      0.68      1294
weighted avg       0.97      0.91      0.93      1294

[[1146  104]
 [   7   37]]

{'f1_score': 0.4,
 'precision': 0.2624113475177305,
 'recall': 0.8409090909090909,
 'accuracy': 0.9142194744976816}


# Initializing dictionary object to store the model performance metrics
feature_selection_tune_GBBoost={}

# Storing the first 50 highest ranked features
features_=list(feat_score['columns'].values[:50])

# Iterating from 50th highest ranked features to last ranked features
for i in feat_score['columns'].values[50:]:
    features_.append(i)
    feature_selection_tune_GBBoost[len(features_)]=get_model_performance(GradientBoostingClassifier()
                                                                    ,data_GBoost,ADASYN(
                                                                        random_state=42,
                                                                        n_neighbors=5,
                                                                        n_jobs=-1),
                                                                    display_report=False,
                                                                    custom_featrues=features_)
pd.DataFrame(feature_selection_tune_GBBoost).T.style.highlight_max()


feature_selection_tune_GBBoost_=pd.DataFrame(feature_selection_tune_GBBoost).T
plot_best_score(feature_selection_tune_GBBoost_,"Tuning by IQR scale for Gradient Boost model")

0.4


# Storing Xtrain set and ytrain set for GradientBoostingClassifier model
X_train=data_rf.drop("Bankrupt?",axis=1)
y_train=data_rf["Bankrupt?"]

# Scalining the data
scaller=StandardScaler()
X_train_=scaller.fit_transform(X_train)
X_test_=scaller.transform(X_test)

# OverSampling the data
X_train_,y_train_=ADASYN(random_state=42,n_neighbors=5,n_jobs=-1).fit_resample(X_train_,y_train)

# Storing the best parameters for tuning GradientBoostingClassifier model
param_grid = {
    "loss":["deviance"],
    "learning_rate": [0.01, 0.025, 0.05, 0.075, 0.1, 0.15, 0.2],
    "min_samples_split": np.linspace(0.1, 0.5, 12),
    "min_samples_leaf": np.linspace(0.1, 0.5, 12),
    "max_depth":[3,5,8],
    "max_features":["log2","sqrt"],
    "criterion": ["friedman_mse",  "mae"],
    "subsample":[0.5, 0.618, 0.8, 0.85, 0.9, 0.95, 1.0],
    "n_estimators":[10,20,30,40,80,100,200]
    }

# Initializing the Random Forest model
model=GradientBoostingClassifier()

# Initializing the Randomised Search CV object
grid=RandomizedSearchCV(estimator=model,param_distributions= param_grid
                        ,cv=5,verbose=10,
                        scoring='neg_mean_absolute_error')

# Fitting the model
grid.fit(X_train_,y_train_)
print(grid.best_params_)

# Storing the best parameters for Random Forest model
model=grid.best_estimator_

# Printing the best parameters for Random Forest model
print_metrics(predictions=model.predict(X_test_))

Fitting 5 folds for each of 10 candidates, totalling 50 fits
[CV 1/5; 1/10] START criterion=mae, learning_rate=0.075, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.390909090909091, min_samples_split=0.5, n_estimators=40, subsample=1.0
[CV 1/5; 1/10] END criterion=mae, learning_rate=0.075, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.390909090909091, min_samples_split=0.5, n_estimators=40, subsample=1.0;, score=-0.217 total time=  18.0s
[CV 2/5; 1/10] START criterion=mae, learning_rate=0.075, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.390909090909091, min_samples_split=0.5, n_estimators=40, subsample=1.0
[CV 2/5; 1/10] END criterion=mae, learning_rate=0.075, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.390909090909091, min_samples_split=0.5, n_estimators=40, subsample=1.0;, score=-0.146 total time=  14.0s
[CV 3/5; 1/10] START criterion=mae, learning_rate=0.075, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.390909090909091, min_samples_split=0.5, n_estimators=40, subsample=1.0
[CV 3/5; 1/10] END criterion=mae, learning_rate=0.075, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.390909090909091, min_samples_split=0.5, n_estimators=40, subsample=1.0;, score=-0.139 total time=  15.8s
[CV 4/5; 1/10] START criterion=mae, learning_rate=0.075, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.390909090909091, min_samples_split=0.5, n_estimators=40, subsample=1.0
[CV 4/5; 1/10] END criterion=mae, learning_rate=0.075, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.390909090909091, min_samples_split=0.5, n_estimators=40, subsample=1.0;, score=-0.196 total time=  14.1s
[CV 5/5; 1/10] START criterion=mae, learning_rate=0.075, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.390909090909091, min_samples_split=0.5, n_estimators=40, subsample=1.0
[CV 5/5; 1/10] END criterion=mae, learning_rate=0.075, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.390909090909091, min_samples_split=0.5, n_estimators=40, subsample=1.0;, score=-0.147 total time=  17.1s
[CV 1/5; 2/10] START criterion=mae, learning_rate=0.2, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.17272727272727273, min_samples_split=0.13636363636363638, n_estimators=20, subsample=0.9
[CV 1/5; 2/10] END criterion=mae, learning_rate=0.2, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.17272727272727273, min_samples_split=0.13636363636363638, n_estimators=20, subsample=0.9;, score=-0.183 total time=   9.4s
[CV 2/5; 2/10] START criterion=mae, learning_rate=0.2, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.17272727272727273, min_samples_split=0.13636363636363638, n_estimators=20, subsample=0.9
[CV 2/5; 2/10] END criterion=mae, learning_rate=0.2, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.17272727272727273, min_samples_split=0.13636363636363638, n_estimators=20, subsample=0.9;, score=-0.135 total time=   8.5s
[CV 3/5; 2/10] START criterion=mae, learning_rate=0.2, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.17272727272727273, min_samples_split=0.13636363636363638, n_estimators=20, subsample=0.9
[CV 3/5; 2/10] END criterion=mae, learning_rate=0.2, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.17272727272727273, min_samples_split=0.13636363636363638, n_estimators=20, subsample=0.9;, score=-0.156 total time=   8.9s
[CV 4/5; 2/10] START criterion=mae, learning_rate=0.2, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.17272727272727273, min_samples_split=0.13636363636363638, n_estimators=20, subsample=0.9
[CV 4/5; 2/10] END criterion=mae, learning_rate=0.2, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.17272727272727273, min_samples_split=0.13636363636363638, n_estimators=20, subsample=0.9;, score=-0.159 total time=   8.9s
[CV 5/5; 2/10] START criterion=mae, learning_rate=0.2, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.17272727272727273, min_samples_split=0.13636363636363638, n_estimators=20, subsample=0.9
[CV 5/5; 2/10] END criterion=mae, learning_rate=0.2, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.17272727272727273, min_samples_split=0.13636363636363638, n_estimators=20, subsample=0.9;, score=-0.147 total time=   8.4s
[CV 1/5; 3/10] START criterion=mae, learning_rate=0.025, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.2090909090909091, min_samples_split=0.42727272727272736, n_estimators=10, subsample=0.8
[CV 1/5; 3/10] END criterion=mae, learning_rate=0.025, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.2090909090909091, min_samples_split=0.42727272727272736, n_estimators=10, subsample=0.8;, score=-0.262 total time=   2.5s
[CV 2/5; 3/10] START criterion=mae, learning_rate=0.025, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.2090909090909091, min_samples_split=0.42727272727272736, n_estimators=10, subsample=0.8
[CV 2/5; 3/10] END criterion=mae, learning_rate=0.025, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.2090909090909091, min_samples_split=0.42727272727272736, n_estimators=10, subsample=0.8;, score=-0.169 total time=   3.3s
[CV 3/5; 3/10] START criterion=mae, learning_rate=0.025, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.2090909090909091, min_samples_split=0.42727272727272736, n_estimators=10, subsample=0.8
[CV 3/5; 3/10] END criterion=mae, learning_rate=0.025, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.2090909090909091, min_samples_split=0.42727272727272736, n_estimators=10, subsample=0.8;, score=-0.211 total time=   3.0s
[CV 4/5; 3/10] START criterion=mae, learning_rate=0.025, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.2090909090909091, min_samples_split=0.42727272727272736, n_estimators=10, subsample=0.8
[CV 4/5; 3/10] END criterion=mae, learning_rate=0.025, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.2090909090909091, min_samples_split=0.42727272727272736, n_estimators=10, subsample=0.8;, score=-0.203 total time=   2.7s
[CV 5/5; 3/10] START criterion=mae, learning_rate=0.025, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.2090909090909091, min_samples_split=0.42727272727272736, n_estimators=10, subsample=0.8
[CV 5/5; 3/10] END criterion=mae, learning_rate=0.025, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.2090909090909091, min_samples_split=0.42727272727272736, n_estimators=10, subsample=0.8;, score=-0.191 total time=   2.9s
[CV 1/5; 4/10] START criterion=friedman_mse, learning_rate=0.075, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.5, min_samples_split=0.5, n_estimators=40, subsample=0.9
[CV 1/5; 4/10] END criterion=friedman_mse, learning_rate=0.075, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.5, min_samples_split=0.5, n_estimators=40, subsample=0.9;, score=-0.498 total time=   0.0s
[CV 2/5; 4/10] START criterion=friedman_mse, learning_rate=0.075, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.5, min_samples_split=0.5, n_estimators=40, subsample=0.9
[CV 2/5; 4/10] END criterion=friedman_mse, learning_rate=0.075, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.5, min_samples_split=0.5, n_estimators=40, subsample=0.9;, score=-0.498 total time=   0.0s
[CV 3/5; 4/10] START criterion=friedman_mse, learning_rate=0.075, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.5, min_samples_split=0.5, n_estimators=40, subsample=0.9
[CV 3/5; 4/10] END criterion=friedman_mse, learning_rate=0.075, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.5, min_samples_split=0.5, n_estimators=40, subsample=0.9;, score=-0.498 total time=   0.0s
[CV 4/5; 4/10] START criterion=friedman_mse, learning_rate=0.075, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.5, min_samples_split=0.5, n_estimators=40, subsample=0.9
[CV 4/5; 4/10] END criterion=friedman_mse, learning_rate=0.075, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.5, min_samples_split=0.5, n_estimators=40, subsample=0.9;, score=-0.498 total time=   0.0s
[CV 5/5; 4/10] START criterion=friedman_mse, learning_rate=0.075, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.5, min_samples_split=0.5, n_estimators=40, subsample=0.9
[CV 5/5; 4/10] END criterion=friedman_mse, learning_rate=0.075, loss=deviance, max_depth=3, max_features=sqrt, min_samples_leaf=0.5, min_samples_split=0.5, n_estimators=40, subsample=0.9;, score=-0.498 total time=   0.0s
[CV 1/5; 5/10] START criterion=friedman_mse, learning_rate=0.025, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.42727272727272736, min_samples_split=0.17272727272727273, n_estimators=100, subsample=0.95
[CV 1/5; 5/10] END criterion=friedman_mse, learning_rate=0.025, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.42727272727272736, min_samples_split=0.17272727272727273, n_estimators=100, subsample=0.95;, score=-0.159 total time=   0.4s
[CV 2/5; 5/10] START criterion=friedman_mse, learning_rate=0.025, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.42727272727272736, min_samples_split=0.17272727272727273, n_estimators=100, subsample=0.95
[CV 2/5; 5/10] END criterion=friedman_mse, learning_rate=0.025, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.42727272727272736, min_samples_split=0.17272727272727273, n_estimators=100, subsample=0.95;, score=-0.120 total time=   0.4s
[CV 3/5; 5/10] START criterion=friedman_mse, learning_rate=0.025, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.42727272727272736, min_samples_split=0.17272727272727273, n_estimators=100, subsample=0.95
[CV 3/5; 5/10] END criterion=friedman_mse, learning_rate=0.025, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.42727272727272736, min_samples_split=0.17272727272727273, n_estimators=100, subsample=0.95;, score=-0.143 total time=   0.4s
[CV 4/5; 5/10] START criterion=friedman_mse, learning_rate=0.025, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.42727272727272736, min_samples_split=0.17272727272727273, n_estimators=100, subsample=0.95
[CV 4/5; 5/10] END criterion=friedman_mse, learning_rate=0.025, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.42727272727272736, min_samples_split=0.17272727272727273, n_estimators=100, subsample=0.95;, score=-0.193 total time=   0.5s
[CV 5/5; 5/10] START criterion=friedman_mse, learning_rate=0.025, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.42727272727272736, min_samples_split=0.17272727272727273, n_estimators=100, subsample=0.95
[CV 5/5; 5/10] END criterion=friedman_mse, learning_rate=0.025, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.42727272727272736, min_samples_split=0.17272727272727273, n_estimators=100, subsample=0.95;, score=-0.145 total time=   0.4s
[CV 1/5; 6/10] START criterion=friedman_mse, learning_rate=0.15, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.46363636363636374, min_samples_split=0.24545454545454548, n_estimators=100, subsample=1.0
[CV 1/5; 6/10] END criterion=friedman_mse, learning_rate=0.15, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.46363636363636374, min_samples_split=0.24545454545454548, n_estimators=100, subsample=1.0;, score=-0.147 total time=   0.4s
[CV 2/5; 6/10] START criterion=friedman_mse, learning_rate=0.15, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.46363636363636374, min_samples_split=0.24545454545454548, n_estimators=100, subsample=1.0
[CV 2/5; 6/10] END criterion=friedman_mse, learning_rate=0.15, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.46363636363636374, min_samples_split=0.24545454545454548, n_estimators=100, subsample=1.0;, score=-0.101 total time=   0.4s
[CV 3/5; 6/10] START criterion=friedman_mse, learning_rate=0.15, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.46363636363636374, min_samples_split=0.24545454545454548, n_estimators=100, subsample=1.0
[CV 3/5; 6/10] END criterion=friedman_mse, learning_rate=0.15, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.46363636363636374, min_samples_split=0.24545454545454548, n_estimators=100, subsample=1.0;, score=-0.134 total time=   0.4s
[CV 4/5; 6/10] START criterion=friedman_mse, learning_rate=0.15, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.46363636363636374, min_samples_split=0.24545454545454548, n_estimators=100, subsample=1.0
[CV 4/5; 6/10] END criterion=friedman_mse, learning_rate=0.15, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.46363636363636374, min_samples_split=0.24545454545454548, n_estimators=100, subsample=1.0;, score=-0.147 total time=   0.4s
[CV 5/5; 6/10] START criterion=friedman_mse, learning_rate=0.15, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.46363636363636374, min_samples_split=0.24545454545454548, n_estimators=100, subsample=1.0
[CV 5/5; 6/10] END criterion=friedman_mse, learning_rate=0.15, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.46363636363636374, min_samples_split=0.24545454545454548, n_estimators=100, subsample=1.0;, score=-0.120 total time=   0.4s
[CV 1/5; 7/10] START criterion=friedman_mse, learning_rate=0.075, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.31818181818181823, min_samples_split=0.24545454545454548, n_estimators=80, subsample=0.85
[CV 1/5; 7/10] END criterion=friedman_mse, learning_rate=0.075, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.31818181818181823, min_samples_split=0.24545454545454548, n_estimators=80, subsample=0.85;, score=-0.153 total time=   0.3s
[CV 2/5; 7/10] START criterion=friedman_mse, learning_rate=0.075, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.31818181818181823, min_samples_split=0.24545454545454548, n_estimators=80, subsample=0.85
[CV 2/5; 7/10] END criterion=friedman_mse, learning_rate=0.075, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.31818181818181823, min_samples_split=0.24545454545454548, n_estimators=80, subsample=0.85;, score=-0.102 total time=   0.3s
[CV 3/5; 7/10] START criterion=friedman_mse, learning_rate=0.075, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.31818181818181823, min_samples_split=0.24545454545454548, n_estimators=80, subsample=0.85
[CV 3/5; 7/10] END criterion=friedman_mse, learning_rate=0.075, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.31818181818181823, min_samples_split=0.24545454545454548, n_estimators=80, subsample=0.85;, score=-0.139 total time=   0.4s
[CV 4/5; 7/10] START criterion=friedman_mse, learning_rate=0.075, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.31818181818181823, min_samples_split=0.24545454545454548, n_estimators=80, subsample=0.85
[CV 4/5; 7/10] END criterion=friedman_mse, learning_rate=0.075, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.31818181818181823, min_samples_split=0.24545454545454548, n_estimators=80, subsample=0.85;, score=-0.161 total time=   0.4s
[CV 5/5; 7/10] START criterion=friedman_mse, learning_rate=0.075, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.31818181818181823, min_samples_split=0.24545454545454548, n_estimators=80, subsample=0.85
[CV 5/5; 7/10] END criterion=friedman_mse, learning_rate=0.075, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.31818181818181823, min_samples_split=0.24545454545454548, n_estimators=80, subsample=0.85;, score=-0.137 total time=   0.5s
[CV 1/5; 8/10] START criterion=friedman_mse, learning_rate=0.1, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.390909090909091, n_estimators=30, subsample=0.618
[CV 1/5; 8/10] END criterion=friedman_mse, learning_rate=0.1, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.390909090909091, n_estimators=30, subsample=0.618;, score=-0.183 total time=   0.3s
[CV 2/5; 8/10] START criterion=friedman_mse, learning_rate=0.1, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.390909090909091, n_estimators=30, subsample=0.618
[CV 2/5; 8/10] END criterion=friedman_mse, learning_rate=0.1, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.390909090909091, n_estimators=30, subsample=0.618;, score=-0.115 total time=   0.2s
[CV 3/5; 8/10] START criterion=friedman_mse, learning_rate=0.1, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.390909090909091, n_estimators=30, subsample=0.618
[CV 3/5; 8/10] END criterion=friedman_mse, learning_rate=0.1, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.390909090909091, n_estimators=30, subsample=0.618;, score=-0.153 total time=   0.1s
[CV 4/5; 8/10] START criterion=friedman_mse, learning_rate=0.1, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.390909090909091, n_estimators=30, subsample=0.618
[CV 4/5; 8/10] END criterion=friedman_mse, learning_rate=0.1, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.390909090909091, n_estimators=30, subsample=0.618;, score=-0.190 total time=   0.1s
[CV 5/5; 8/10] START criterion=friedman_mse, learning_rate=0.1, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.390909090909091, n_estimators=30, subsample=0.618
[CV 5/5; 8/10] END criterion=friedman_mse, learning_rate=0.1, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.390909090909091, n_estimators=30, subsample=0.618;, score=-0.140 total time=   0.1s
[CV 1/5; 9/10] START criterion=mae, learning_rate=0.15, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.28181818181818186, n_estimators=30, subsample=1.0
[CV 1/5; 9/10] END criterion=mae, learning_rate=0.15, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.28181818181818186, n_estimators=30, subsample=1.0;, score=-0.200 total time=  21.2s
[CV 2/5; 9/10] START criterion=mae, learning_rate=0.15, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.28181818181818186, n_estimators=30, subsample=1.0
[CV 2/5; 9/10] END criterion=mae, learning_rate=0.15, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.28181818181818186, n_estimators=30, subsample=1.0;, score=-0.116 total time=  15.8s
[CV 3/5; 9/10] START criterion=mae, learning_rate=0.15, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.28181818181818186, n_estimators=30, subsample=1.0
[CV 3/5; 9/10] END criterion=mae, learning_rate=0.15, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.28181818181818186, n_estimators=30, subsample=1.0;, score=-0.151 total time=  18.8s
[CV 4/5; 9/10] START criterion=mae, learning_rate=0.15, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.28181818181818186, n_estimators=30, subsample=1.0
[CV 4/5; 9/10] END criterion=mae, learning_rate=0.15, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.28181818181818186, n_estimators=30, subsample=1.0;, score=-0.173 total time=  15.8s
[CV 5/5; 9/10] START criterion=mae, learning_rate=0.15, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.28181818181818186, n_estimators=30, subsample=1.0
[CV 5/5; 9/10] END criterion=mae, learning_rate=0.15, loss=deviance, max_depth=5, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.28181818181818186, n_estimators=30, subsample=1.0;, score=-0.167 total time=  15.1s
[CV 1/5; 10/10] START criterion=friedman_mse, learning_rate=0.025, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.2090909090909091, n_estimators=10, subsample=1.0
[CV 1/5; 10/10] END criterion=friedman_mse, learning_rate=0.025, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.2090909090909091, n_estimators=10, subsample=1.0;, score=-0.225 total time=   0.0s
[CV 2/5; 10/10] START criterion=friedman_mse, learning_rate=0.025, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.2090909090909091, n_estimators=10, subsample=1.0
[CV 2/5; 10/10] END criterion=friedman_mse, learning_rate=0.025, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.2090909090909091, n_estimators=10, subsample=1.0;, score=-0.144 total time=   0.0s
[CV 3/5; 10/10] START criterion=friedman_mse, learning_rate=0.025, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.2090909090909091, n_estimators=10, subsample=1.0
[CV 3/5; 10/10] END criterion=friedman_mse, learning_rate=0.025, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.2090909090909091, n_estimators=10, subsample=1.0;, score=-0.161 total time=   0.0s
[CV 4/5; 10/10] START criterion=friedman_mse, learning_rate=0.025, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.2090909090909091, n_estimators=10, subsample=1.0
[CV 4/5; 10/10] END criterion=friedman_mse, learning_rate=0.025, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.2090909090909091, n_estimators=10, subsample=1.0;, score=-0.172 total time=   0.0s
[CV 5/5; 10/10] START criterion=friedman_mse, learning_rate=0.025, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.2090909090909091, n_estimators=10, subsample=1.0
[CV 5/5; 10/10] END criterion=friedman_mse, learning_rate=0.025, loss=deviance, max_depth=3, max_features=log2, min_samples_leaf=0.24545454545454548, min_samples_split=0.2090909090909091, n_estimators=10, subsample=1.0;, score=-0.180 total time=   0.0s
{'subsample': 1.0, 'n_estimators': 100, 'min_samples_split': 0.24545454545454548, 'min_samples_leaf': 0.46363636363636374, 'max_features': 'log2', 'max_depth': 3, 'loss': 'deviance', 'learning_rate': 0.15, 'criterion': 'friedman_mse'}
0.28671328671328666 f1score
0.8423493044822257 accuracy
0.16942148760330578 precision
0.9318181818181818 recall
              precision    recall  f1-score   support

           0       1.00      0.84      0.91      1250
           1       0.17      0.93      0.29        44

    accuracy                           0.84      1294
   macro avg       0.58      0.89      0.60      1294
weighted avg       0.97      0.84      0.89      1294

[[1049  201]
 [   3   41]]

{'f1_score': 0.28671328671328666,
 'precision': 0.16942148760330578,
 'recall': 0.9318181818181818,
 'accuracy': 0.8423493044822257}


# Data _suiteable for XGBoost Model
IQR_scale_xgb=3.8
data_xgb_=impute_outliers(bank_data_stage_2,IQR_scale_xgb,list(ascendingly_order_outlier_occr_feat[:33]))
#
# Storing the first 52 ranked features
features_for_XgBoost_=list(feat_score['columns'].values[:52])

# Storing over Sampling method ADASYN
over_sampling_method=ADASYN(random_state=42,n_neighbors=5,n_jobs=-1)

# model
xgboost=XGBClassifier()


xgb_model_metrics,xgb_model,scalled_test_set_xgb=get_model_performance(xgboost,data_xgb_,
                                                          over_sampling_method,
                                                          display_report=True,
                                                          custom_featrues=features_for_XgBoost_,
                                                          return_trained_model=True)
y_pred_proba_xgb=xgb_model.predict_proba(scalled_test_set_xgb)[:,1]

precision_xgb, recall_xgb, _xgb = precision_recall_curve(y_test, y_pred_proba_xgb)
fpr_xgb, tpr_xgb, _xgb = roc_curve(y_test, y_pred_proba_xgb, pos_label=1)
print("AUC SCORE OF XGBOOST",roc_auc_score(y_test,y_pred_proba_xgb))
cm_xgb=confusion_matrix(y_test,xgb_model.predict(scalled_test_set_xgb))
plt.figure(figsize=(3,3),dpi=200)
ax=sns.heatmap(cm_xgb,annot=True,fmt='g');
ax.set_title("Confusion Matrix XGBoost")
ax.set_xlabel('Predicted Labels')
ax.set_ylabel('Actual Labels')
ax.xaxis.set_ticklabels(['Not Bankrupt',"Bankrupted"])
ax.yaxis.set_ticklabels(['Not Bankrupt',"Bankrupted"])
plt.show()

[19:01:39] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.5.1/src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
0.5454545454545454 f1score
0.9574961360123647 accuracy
0.42857142857142855 precision
0.75 recall
              precision    recall  f1-score   support

           0       0.99      0.96      0.98      1250
           1       0.43      0.75      0.55        44

    accuracy                           0.96      1294
   macro avg       0.71      0.86      0.76      1294
weighted avg       0.97      0.96      0.96      1294

[[1206   44]
 [  11   33]]
AUC SCORE OF XGBOOST 0.9434727272727272


IQR_scale_rf=2.5
data_rf_=impute_outliers(bank_data_stage_2,IQR_scale,list(ascendingly_order_outlier_occr_feat[:33]))
features_for_Rf_=list(feat_score['columns'].values[:69])

rf_model_metrics,rf_model,scalled_test_set_rf=get_model_performance(RandomForestClassifier(),
                      data_rf,ADASYN(random_state=42,n_neighbors=5,n_jobs=-1)
                      ,display_report=True,
                      custom_featrues=features_for_Rf_,
                        return_trained_model=True)

y_pred_proba_rf=rf_model.predict_proba(scalled_test_set_rf)[:,1]
precision_rf, recall_rf, _rf = precision_recall_curve(y_test, y_pred_proba_rf)
fpr_rf, tpr_rf, _rf = roc_curve(y_test, y_pred_proba_rf, pos_label=1)
print("AUC SCORE OF Random Forest",roc_auc_score(y_test,y_pred_proba_rf))


plt.figure(figsize=(3,3),dpi=200)
cm_rf=confusion_matrix(y_test,rf_model.predict(scalled_test_set_rf))
ax=sns.heatmap(cm_rf,annot=True,fmt='g');
ax.set_title("Confusion Matrix XGBoost")
ax.set_xlabel('Predicted Labels')
ax.set_ylabel('Actual Labels')
ax.xaxis.set_ticklabels(['Not Bankrupt',"Bankrupted"])
ax.yaxis.set_ticklabels(['Not Bankrupt',"Bankrupted"])
plt.show()

0.45945945945945943 f1score
0.9381761978361669 accuracy
0.3269230769230769 precision
0.7727272727272727 recall
              precision    recall  f1-score   support

           0       0.99      0.94      0.97      1250
           1       0.33      0.77      0.46        44

    accuracy                           0.94      1294
   macro avg       0.66      0.86      0.71      1294
weighted avg       0.97      0.94      0.95      1294

[[1180   70]
 [  10   34]]
AUC SCORE OF Random Forest 0.9587454545454546


plt.style.use('seaborn')

# plot roc curves
plt.plot(fpr_xgb, tpr_xgb, linestyle='--',color='orange', label='XGBoost')
plt.plot(fpr_rf, tpr_rf, linestyle='--',color='green', label='Random Forest')
# title
plt.title('ROC curve')
# x label
plt.xlabel('False Positive Rate')
# y label
plt.ylabel('True Positive rate')

plt.legend(loc='best')
plt.savefig('ROC',dpi=300)
plt.show();


plt.figure(figsize=(10,5),dpi=200)
plt.plot(precision_xgb, recall_xgb,color='purple',label="XGBoost")
plt.plot(precision_rf,recall_rf,  color='red',label="Random Forest")
#add axis labels to plot
plt.title('Precision-Recall Curve')
plt.xlabel('Precision')
plt.ylabel('Recall')

# Highlighting the best part of curve
plt.plot(0.3402,0.75,label="Best Random Forest Precision Recall Trade off", markersize=15, markeredgecolor="red", markerfacecolor="green",marker="o",)
plt.plot(0.42857,0.75,label="Best XGBoost Precision Recall Trade off", markersize=15, markeredgecolor="red", markerfacecolor="black",marker="o",)

plt.legend()
plt.savefig('PRC',dpi=300)
plt.show()

	Bankrupt?	ROA(C) before interest and depreciation before interest	ROA(A) before interest and % after tax	ROA(B) before interest and depreciation after tax	Operating Gross Margin	Realized Sales Gross Margin	Operating Profit Rate	Pre-tax net Interest Rate	After-tax net Interest Rate	Non-industry income and expenditure/revenue	Continuous interest rate (after tax)	Operating Expense Rate	Research and development expense rate	Cash flow rate	Interest-bearing debt interest rate	Net Value Per Share (B)	Net Value Per Share (A)	Net Value Per Share (C)	Persistent EPS in the Last Four Seasons	Cash Flow Per Share	Revenue Per Share (Yuan ¥)	Operating Profit Per Share (Yuan ¥)	Per Share Net profit before tax (Yuan ¥)	Realized Sales Gross Profit Growth Rate	Operating Profit Growth Rate	After-tax Net Profit Growth Rate	Regular Net Profit Growth Rate	Continuous Net Profit Growth Rate	Total Asset Growth Rate	Net Value Growth Rate	Total Asset Return Growth Rate Ratio	Cash Reinvestment %	Current Ratio	Quick Ratio	Interest Expense Ratio	Total debt/Total net worth	Debt ratio %	Net worth/Assets	Long-term fund suitability ratio (A)	Borrowing dependency	Contingent liabilities/Net worth	Operating profit/Paid-in capital	Net profit before tax/Paid-in capital	Inventory and accounts receivable/Net value	Total Asset Turnover	Accounts Receivable Turnover	Average Collection Days	Inventory Turnover Rate (times)	Fixed Assets Turnover Frequency	Net Worth Turnover Rate (times)	Revenue per person	Operating profit per person	Allocation rate per person	Working Capital to Total Assets	Quick Assets/Total Assets	Current Assets/Total Assets	Cash/Total Assets	Quick Assets/Current Liability	Cash/Current Liability	Current Liability to Assets	Operating Funds to Liability	Inventory/Working Capital	Inventory/Current Liability	Current Liabilities/Liability	Working Capital/Equity	Current Liabilities/Equity	Long-term Liability to Current Assets	Retained Earnings to Total Assets	Total income/Total expense	Total expense/Assets	Current Asset Turnover Rate	Quick Asset Turnover Rate	Working capitcal Turnover Rate	Cash Turnover Rate	Cash Flow to Sales	Fixed Assets to Assets	Current Liability to Liability	Current Liability to Equity	Equity to Long-term Liability	Cash Flow to Total Assets	Cash Flow to Liability	CFO to Assets	Cash Flow to Equity	Current Liability to Current Assets	Net Income to Total Assets	Total assets to GNP price	No-credit Interval	Gross Profit to Sales	Net Income to Stockholder's Equity	Liability to Equity	Degree of Financial Leverage (DFL)	Interest Coverage Ratio (Interest expense to EBIT)	Net Income Flag	Equity to Liability
0	1	0.370594	0.424389	0.405750	0.601457	0.601457	0.998969	0.796887	0.808809	0.302646	0.780985	1.256969e-04	0.0	0.458143	0.000725	0.147950	0.147950	0.147950	0.169141	0.311664	0.017560	0.095921	0.138736	0.022102	0.848195	0.688979	0.688979	0.217535	4.980000e+09	0.000327	0.263100	0.363725	0.002259	0.001208	0.629951	0.021266	0.207576	0.792424	0.005024	0.390284	0.006479	0.095885	0.137757	0.398036	0.086957	0.001814	0.003487	1.820926e-04	1.165007e-04	0.032903	0.034164	0.392913	0.037135	0.672775	0.166673	0.190643	0.004094	0.001997	1.473360e-04	0.147308	0.334015	0.276920	0.001036	0.676269	0.721275	0.339077	0.025592	0.903225	0.002022	0.064856	7.010000e+08	6.550000e+09	0.593831	4.580000e+08	0.671568	0.424206	0.676269	0.339077	0.126549	0.637555	0.458609	0.520382	0.312905	0.118250	0.716845	0.009219	0.622879	0.601453	0.827890	0.290202	0.026601	0.564050	1	0.016469
1	1	0.464291	0.538214	0.516730	0.610235	0.610235	0.998946	0.797380	0.809301	0.303556	0.781506	2.897851e-04	0.0	0.461867	0.000647	0.182251	0.182251	0.182251	0.208944	0.318137	0.021144	0.093722	0.169918	0.022080	0.848088	0.689693	0.689702	0.217620	6.110000e+09	0.000443	0.264516	0.376709	0.006016	0.004039	0.635172	0.012502	0.171176	0.828824	0.005059	0.376760	0.005835	0.093743	0.168962	0.397725	0.064468	0.001286	0.004917	9.360000e+09	7.190000e+08	0.025484	0.006889	0.391590	0.012335	0.751111	0.127236	0.182419	0.014948	0.004136	1.383910e-03	0.056963	0.341106	0.289642	0.005210	0.308589	0.731975	0.329740	0.023947	0.931065	0.002226	0.025516	1.065198e-04	7.700000e+09	0.593916	2.490000e+09	0.671570	0.468828	0.308589	0.329740	0.120916	0.641100	0.459001	0.567101	0.314163	0.047775	0.795297	0.008323	0.623652	0.610237	0.839969	0.283846	0.264577	0.570175	1	0.020794
2	1	0.426071	0.499019	0.472295	0.601450	0.601364	0.998857	0.796403	0.808388	0.302035	0.780284	2.361297e-04	25500000.0	0.458521	0.000790	0.177911	0.177911	0.193713	0.180581	0.307102	0.005944	0.092338	0.142803	0.022760	0.848094	0.689463	0.689470	0.217601	7.280000e+09	0.000396	0.264184	0.368913	0.011543	0.005348	0.629631	0.021248	0.207516	0.792484	0.005100	0.379093	0.006562	0.092318	0.148036	0.406580	0.014993	0.001495	0.004227	6.500000e+07	2.650000e+09	0.013387	0.028997	0.381968	0.141016	0.829502	0.340201	0.602806	0.000991	0.006302	5.340000e+09	0.098162	0.336731	0.277456	0.013879	0.446027	0.742729	0.334777	0.003715	0.909903	0.002060	0.021387	1.791094e-03	1.022676e-03	0.594502	7.610000e+08	0.671571	0.276179	0.446027	0.334777	0.117922	0.642765	0.459254	0.538491	0.314515	0.025346	0.774670	0.040003	0.623841	0.601449	0.836774	0.290189	0.026555	0.563706	1	0.016474
3	1	0.399844	0.451265	0.457733	0.583541	0.583541	0.998700	0.796967	0.808966	0.303350	0.781241	1.078888e-04	0.0	0.465705	0.000449	0.154187	0.154187	0.154187	0.193722	0.321674	0.014368	0.077762	0.148603	0.022046	0.848005	0.689110	0.689110	0.217568	4.880000e+09	0.000382	0.263371	0.384077	0.004194	0.002896	0.630228	0.009572	0.151465	0.848535	0.005047	0.379743	0.005366	0.077727	0.147561	0.397925	0.089955	0.001966	0.003215	7.130000e+09	9.150000e+09	0.028065	0.015463	0.378497	0.021320	0.725754	0.161575	0.225815	0.018851	0.002961	1.010646e-03	0.098715	0.348716	0.276580	0.003540	0.615848	0.729825	0.331509	0.022165	0.906902	0.001831	0.024161	8.140000e+09	6.050000e+09	0.593889	2.030000e+09	0.671519	0.559144	0.615848	0.331509	0.120760	0.579039	0.448518	0.604105	0.302382	0.067250	0.739555	0.003252	0.622929	0.583538	0.834697	0.281721	0.026697	0.564663	1	0.023982
4	1	0.465022	0.538432	0.522298	0.598783	0.598783	0.998973	0.797366	0.809304	0.303475	0.781550	7.890000e+09	0.0	0.462746	0.000686	0.167502	0.167502	0.167502	0.212537	0.319162	0.029690	0.096898	0.168412	0.022096	0.848258	0.689697	0.689697	0.217626	5.510000e+09	0.000439	0.265218	0.379690	0.006022	0.003727	0.636055	0.005150	0.106509	0.893491	0.005303	0.375025	0.006624	0.096927	0.167461	0.400079	0.175412	0.001449	0.004367	1.633674e-04	2.935211e-04	0.040161	0.058111	0.394371	0.023988	0.751822	0.260330	0.358380	0.014161	0.004275	6.804636e-04	0.110195	0.344639	0.287913	0.004869	0.975007	0.732000	0.330726	0.000000	0.913850	0.002224	0.026385	6.680000e+09	5.050000e+09	0.593915	8.240000e+08	0.671563	0.309555	0.975007	0.330726	0.110933	0.622374	0.454411	0.578469	0.311567	0.047725	0.795016	0.003878	0.623521	0.598782	0.839973	0.278514	0.024752	0.575617	1	0.035490

	ROA(C) before interest and depreciation before interest	ROA(A) before interest and % after tax	ROA(B) before interest and depreciation after tax	Operating Gross Margin	Realized Sales Gross Margin	Operating Profit Rate	Pre-tax net Interest Rate	After-tax net Interest Rate	Non-industry income and expenditure/revenue	Continuous interest rate (after tax)	Operating Expense Rate	Research and development expense rate	Cash flow rate	Interest-bearing debt interest rate	Tax rate (A)	Net Value Per Share (B)	Net Value Per Share (A)	Net Value Per Share (C)	Persistent EPS in the Last Four Seasons	Cash Flow Per Share	Revenue Per Share (Yuan ¥)	Operating Profit Per Share (Yuan ¥)	Per Share Net profit before tax (Yuan ¥)	Realized Sales Gross Profit Growth Rate	Operating Profit Growth Rate	After-tax Net Profit Growth Rate	Regular Net Profit Growth Rate	Continuous Net Profit Growth Rate	Total Asset Growth Rate	Net Value Growth Rate	Total Asset Return Growth Rate Ratio	Cash Reinvestment %	Current Ratio	Quick Ratio	Interest Expense Ratio	Total debt/Total net worth	Debt ratio %	Net worth/Assets	Long-term fund suitability ratio (A)	Borrowing dependency	Contingent liabilities/Net worth	Operating profit/Paid-in capital	Net profit before tax/Paid-in capital	Inventory and accounts receivable/Net value	Total Asset Turnover	Accounts Receivable Turnover	Average Collection Days	Inventory Turnover Rate (times)	Fixed Assets Turnover Frequency	Net Worth Turnover Rate (times)	Revenue per person	Operating profit per person	Allocation rate per person	Working Capital to Total Assets	Quick Assets/Total Assets	Current Assets/Total Assets	Cash/Total Assets	Quick Assets/Current Liability	Cash/Current Liability	Current Liability to Assets	Operating Funds to Liability	Inventory/Working Capital	Inventory/Current Liability	Current Liabilities/Liability	Working Capital/Equity	Current Liabilities/Equity	Long-term Liability to Current Assets	Retained Earnings to Total Assets	Total income/Total expense	Total expense/Assets	Current Asset Turnover Rate	Quick Asset Turnover Rate	Working capitcal Turnover Rate	Cash Turnover Rate	Cash Flow to Sales	Fixed Assets to Assets	Current Liability to Liability	Current Liability to Equity	Equity to Long-term Liability	Cash Flow to Total Assets	Cash Flow to Liability	CFO to Assets	Cash Flow to Equity	Current Liability to Current Assets	Net Income to Total Assets	Total assets to GNP price	No-credit Interval	Gross Profit to Sales	Net Income to Stockholder's Equity	Liability to Equity	Degree of Financial Leverage (DFL)	Interest Coverage Ratio (Interest expense to EBIT)	Net Income Flag	Equity to Liability
6814	0.493687	0.539468	0.543230	0.604455	0.604462	0.998992	0.797409	0.809331	0.303510	0.781588	1.510213e-04	4.500000e+09	0.463734	1.790179e-04	0.113372	0.175045	0.175045	0.175045	0.216602	0.320966	0.020766	0.098200	0.172102	0.022374	0.848205	0.689778	0.689778	0.217635	7.070000e+09	0.000450	0.264517	0.380155	0.010451	0.005457	0.631415	0.006655	0.124618	0.875382	0.005150	0.373823	0.005366	0.098222	0.171111	0.404804	0.103448	0.000690	0.009177	4.030000e+07	0.000143	0.027903	0.006348	0.392596	0.006312	0.817769	0.312840	0.578455	0.099481	0.005469	0.005072	0.103838	0.346224	0.277543	0.013212	0.786888	0.736716	0.330914	1.792237e-03	0.925611	0.002266	0.019060	0.000229	0.000124	0.593985	1.077940e-04	0.671570	0.400338	0.786888	0.330914	0.112622	0.639806	0.458639	0.587178	0.314063	0.027951	0.799927	0.000466	0.623620	0.604455	0.840359	0.279606	0.027064	0.566193	1	0.029890
6815	0.475162	0.538269	0.524172	0.598308	0.598308	0.998992	0.797414	0.809327	0.303520	0.781586	5.220000e+09	1.440000e+09	0.461978	2.370237e-04	0.371596	0.181324	0.181324	0.181324	0.216697	0.318278	0.023050	0.098608	0.172780	0.022159	0.848245	0.689734	0.689734	0.217631	5.220000e+09	0.000445	0.264730	0.377389	0.009259	0.006741	0.631489	0.004623	0.099253	0.900747	0.006772	0.372505	0.008619	0.098572	0.171805	0.399926	0.103448	0.000655	0.009652	9.940000e+09	0.000605	0.027419	0.016083	0.393625	0.003401	0.793387	0.335085	0.444043	0.080337	0.006790	0.004727	0.089901	0.342166	0.277368	0.006730	0.849898	0.734584	0.329753	2.204673e-03	0.932629	0.002288	0.011118	0.000152	0.000117	0.593954	7.710000e+09	0.671572	0.096136	0.849898	0.329753	0.112329	0.642072	0.459058	0.569498	0.314446	0.031470	0.799748	0.001959	0.623931	0.598306	0.840306	0.278132	0.027009	0.566018	1	0.038284
6816	0.472725	0.533744	0.520638	0.610444	0.610213	0.998984	0.797401	0.809317	0.303512	0.781546	2.509312e-04	1.039086e-04	0.472189	0.000000e+00	0.490839	0.269521	0.269521	0.269521	0.210929	0.324857	0.044255	0.100073	0.173232	0.022068	0.847978	0.689202	0.689202	0.217547	5.990000e+09	0.000435	0.263858	0.379392	0.038424	0.035112	0.630612	0.001392	0.038939	0.961061	0.009149	0.369637	0.005366	0.100103	0.172287	0.395592	0.106447	0.001510	0.004188	2.797309e-04	0.001024	0.022419	0.022097	0.393693	0.002774	0.866047	0.476747	0.496053	0.412885	0.035531	0.088212	0.024414	0.358847	0.277022	0.007810	0.553964	0.737432	0.326921	0.000000e+00	0.932000	0.002239	0.035446	0.000176	0.000175	0.594025	4.074263e-04	0.671564	0.055509	0.553964	0.326921	0.110933	0.631678	0.452465	0.589341	0.313353	0.007542	0.797778	0.002840	0.624156	0.610441	0.840138	0.275789	0.026791	0.565158	1	0.097649
6817	0.506264	0.559911	0.554045	0.607850	0.607850	0.999074	0.797500	0.809399	0.303498	0.781663	1.236154e-04	2.510000e+09	0.476123	2.110211e-04	0.181294	0.213392	0.213392	0.213392	0.228326	0.346573	0.031535	0.111799	0.185584	0.022350	0.854064	0.696113	0.696113	0.218006	7.250000e+09	0.000529	0.264409	0.401028	0.012782	0.007256	0.630731	0.003816	0.086979	0.913021	0.005529	0.369649	0.007068	0.111722	0.182498	0.401540	0.109445	0.000716	0.008829	4.550000e+09	0.000233	0.027258	0.012749	0.396735	0.007489	0.832340	0.353624	0.564439	0.112238	0.007753	0.007133	0.083199	0.380251	0.277353	0.013334	0.893241	0.736713	0.329294	3.200000e+09	0.939613	0.002395	0.016443	0.000214	0.000135	0.593997	1.165392e-04	0.671606	0.246805	0.893241	0.329294	0.110957	0.684857	0.471313	0.678338	0.320118	0.022916	0.811808	0.002837	0.623957	0.607846	0.841084	0.277547	0.026822	0.565302	1	0.044009
6818	0.493053	0.570105	0.549548	0.627409	0.627409	0.998080	0.801987	0.813800	0.313415	0.786079	1.431695e-03	0.000000e+00	0.427721	5.900000e+08	0.000000	0.220766	0.220766	0.220766	0.227758	0.305793	0.000665	0.092501	0.182119	0.025316	0.848053	0.689527	0.689527	0.217605	9.350000e+09	0.000519	0.264186	0.360102	0.051348	0.040897	0.630618	0.000461	0.014149	0.985851	0.058476	0.370049	0.006368	0.092465	0.179911	0.393883	0.002999	0.000325	0.019474	1.910000e+07	0.000300	0.009194	0.002097	0.385767	0.000963	0.873759	0.527136	0.505010	0.238147	0.051481	0.066674	0.018517	0.239585	0.276975	0.000000	1.000000	0.737286	0.326690	0.000000e+00	0.938005	0.002791	0.006089	0.007864	0.008238	0.598674	9.505992e-03	0.672096	0.005016	1.000000	0.326690	0.110933	0.659917	0.483285	0.505531	0.316238	0.005579	0.815956	0.000707	0.626680	0.627408	0.841019	0.275114	0.026793	0.565167	1	0.233902

	Bankrupt?	ROA(C) before interest and depreciation before interest	ROA(A) before interest and % after tax	ROA(B) before interest and depreciation after tax	Operating Gross Margin	Realized Sales Gross Margin	Operating Profit Rate	Pre-tax net Interest Rate	After-tax net Interest Rate	Non-industry income and expenditure/revenue	Continuous interest rate (after tax)	Operating Expense Rate	Research and development expense rate	Cash flow rate	Interest-bearing debt interest rate	Tax rate (A)	Net Value Per Share (B)	Net Value Per Share (A)	Net Value Per Share (C)	Persistent EPS in the Last Four Seasons	Cash Flow Per Share	Revenue Per Share (Yuan ¥)	Operating Profit Per Share (Yuan ¥)	Per Share Net profit before tax (Yuan ¥)	Realized Sales Gross Profit Growth Rate	Operating Profit Growth Rate	After-tax Net Profit Growth Rate	Regular Net Profit Growth Rate	Continuous Net Profit Growth Rate	Total Asset Growth Rate	Net Value Growth Rate	Total Asset Return Growth Rate Ratio	Cash Reinvestment %	Current Ratio	Quick Ratio	Interest Expense Ratio	Total debt/Total net worth	Debt ratio %	Net worth/Assets	Long-term fund suitability ratio (A)	Borrowing dependency	Contingent liabilities/Net worth	Operating profit/Paid-in capital	Net profit before tax/Paid-in capital	Inventory and accounts receivable/Net value	Total Asset Turnover	Accounts Receivable Turnover	Average Collection Days	Inventory Turnover Rate (times)	Fixed Assets Turnover Frequency	Net Worth Turnover Rate (times)	Revenue per person	Operating profit per person	Allocation rate per person	Working Capital to Total Assets	Quick Assets/Total Assets	Current Assets/Total Assets	Cash/Total Assets	Quick Assets/Current Liability	Cash/Current Liability	Current Liability to Assets	Operating Funds to Liability	Inventory/Working Capital	Inventory/Current Liability	Current Liabilities/Liability	Working Capital/Equity	Current Liabilities/Equity	Long-term Liability to Current Assets	Retained Earnings to Total Assets	Total income/Total expense	Total expense/Assets	Current Asset Turnover Rate	Quick Asset Turnover Rate	Working capitcal Turnover Rate	Cash Turnover Rate	Cash Flow to Sales	Fixed Assets to Assets	Current Liability to Liability	Current Liability to Equity	Equity to Long-term Liability	Cash Flow to Total Assets	Cash Flow to Liability	CFO to Assets	Cash Flow to Equity	Current Liability to Current Assets	Liability-Assets Flag	Net Income to Total Assets	Total assets to GNP price	No-credit Interval	Gross Profit to Sales	Net Income to Stockholder's Equity	Liability to Equity	Degree of Financial Leverage (DFL)	Interest Coverage Ratio (Interest expense to EBIT)	Net Income Flag	Equity to Liability
count	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6.819000e+03	6.819000e+03	6819.000000	6.819000e+03	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6.819000e+03	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6.819000e+03	6.819000e+03	6819.000000	6819.000000	6.819000e+03	6.819000e+03	6819.000000	6.819000e+03	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6.819000e+03	6.819000e+03	6.819000e+03	6.819000e+03	6819.000000	6.819000e+03	6819.000000	6.819000e+03	6819.000000	6819.000000	6819.000000	6819.000000	6.819000e+03	6.819000e+03	6819.000000	6819.000000	6819.000000	6.819000e+03	6819.000000	6819.000000	6819.000000	6.819000e+03	6819.000000	6819.000000	6819.000000	6.819000e+03	6.819000e+03	6819.000000	6.819000e+03	6819.000000	6.819000e+03	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6.819000e+03	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6819.000000	6819.0	6819.000000
mean	0.032263	0.505180	0.558625	0.553589	0.607948	0.607929	0.998755	0.797190	0.809084	0.303623	0.781381	1.995347e+09	1.950427e+09	0.467431	1.644801e+07	0.115001	0.190661	0.190633	0.190672	0.228813	0.323482	1.328641e+06	0.109091	0.184361	0.022408	0.847980	0.689146	0.689150	0.217639	5.508097e+09	1.566212e+06	0.264248	0.379677	4.032850e+05	8.376595e+06	0.630991	4.416337e+06	0.113177	0.886823	0.008783	0.374654	0.005968	0.108977	0.182715	0.402459	0.141606	1.278971e+07	9.826221e+06	2.149106e+09	1.008596e+09	0.038595	2.325854e+06	0.400671	1.125579e+07	0.814125	0.400132	0.522273	0.124095	3.592902e+06	3.715999e+07	0.090673	0.353828	0.277395	5.580680e+07	0.761599	0.735817	0.331410	5.416004e+07	0.934733	0.002549	0.029184	1.195856e+09	2.163735e+09	0.594006	2.471977e+09	0.671531	1.220121e+06	0.761599	0.331410	0.115645	0.649731	0.461849	0.593415	0.315582	0.031506	0.001173	0.807760	1.862942e+07	0.623915	0.607946	0.840402	0.280365	0.027541	0.565358	1.0	0.047578
std	0.176710	0.060686	0.065620	0.061595	0.016934	0.016916	0.013010	0.012869	0.013601	0.011163	0.012679	3.237684e+09	2.598292e+09	0.017036	1.082750e+08	0.138667	0.033390	0.033474	0.033480	0.033263	0.017611	5.170709e+07	0.027942	0.033180	0.012079	0.010752	0.013853	0.013910	0.010063	2.897718e+09	1.141594e+08	0.009634	0.020737	3.330216e+07	2.446847e+08	0.011238	1.684069e+08	0.053920	0.053920	0.028153	0.016286	0.012188	0.027782	0.030785	0.013324	0.101145	2.782598e+08	2.563589e+08	3.247967e+09	2.477557e+09	0.036680	1.366327e+08	0.032720	2.945063e+08	0.059054	0.201998	0.218112	0.139251	1.716209e+08	5.103509e+08	0.050290	0.035147	0.010469	5.820516e+08	0.206677	0.011678	0.013488	5.702706e+08	0.025564	0.012093	0.027149	2.821161e+09	3.374944e+09	0.008959	2.938623e+09	0.009341	1.007542e+08	0.206677	0.013488	0.019529	0.047372	0.029943	0.058561	0.012961	0.030845	0.034234	0.040332	3.764501e+08	0.012290	0.016934	0.014523	0.014463	0.015668	0.013214	0.0	0.050014
min	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000e+00	0.000000e+00	0.000000	0.000000e+00	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000e+00	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000e+00	0.000000e+00	0.000000	0.000000	0.000000e+00	0.000000e+00	0.000000	0.000000e+00	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000e+00	0.000000e+00	0.000000e+00	0.000000e+00	0.000000	0.000000e+00	0.000000	0.000000e+00	0.000000	0.000000	0.000000	0.000000	0.000000e+00	0.000000e+00	0.000000	0.000000	0.000000	0.000000e+00	0.000000	0.000000	0.000000	0.000000e+00	0.000000	0.000000	0.000000	0.000000e+00	0.000000e+00	0.000000	0.000000e+00	0.000000	0.000000e+00	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000e+00	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	1.0	0.000000
25%	0.000000	0.476527	0.535543	0.527277	0.600445	0.600434	0.998969	0.797386	0.809312	0.303466	0.781567	1.566874e-04	1.281880e-04	0.461558	2.030203e-04	0.000000	0.173613	0.173613	0.173676	0.214711	0.317748	1.563138e-02	0.096083	0.170370	0.022065	0.847984	0.689270	0.689270	0.217580	4.860000e+09	4.409689e-04	0.263759	0.374749	7.555047e-03	4.725903e-03	0.630612	3.007049e-03	0.072891	0.851196	0.005244	0.370168	0.005366	0.096105	0.169376	0.397403	0.076462	7.101336e-04	4.386530e-03	1.728256e-04	2.330013e-04	0.021774	1.043285e-02	0.392438	4.120529e-03	0.774309	0.241973	0.352845	0.033543	5.239776e-03	1.973008e-03	0.053301	0.341023	0.277034	3.163148e-03	0.626981	0.733612	0.328096	0.000000e+00	0.931097	0.002236	0.014567	1.456236e-04	1.417149e-04	0.593934	2.735337e-04	0.671565	8.536037e-02	0.626981	0.328096	0.110933	0.633265	0.457116	0.565987	0.312995	0.018034	0.000000	0.796750	9.036205e-04	0.623636	0.600443	0.840115	0.276944	0.026791	0.565158	1.0	0.024477
50%	0.000000	0.502706	0.559802	0.552278	0.605997	0.605976	0.999022	0.797464	0.809375	0.303525	0.781635	2.777589e-04	5.090000e+08	0.465080	3.210321e-04	0.073489	0.184400	0.184400	0.184400	0.224544	0.322487	2.737571e-02	0.104226	0.179709	0.022102	0.848044	0.689439	0.689439	0.217598	6.400000e+09	4.619555e-04	0.264050	0.380425	1.058717e-02	7.412472e-03	0.630698	5.546284e-03	0.111407	0.888593	0.005665	0.372624	0.005366	0.104133	0.178456	0.400131	0.118441	9.678107e-04	6.572537e-03	7.646743e-04	5.930942e-04	0.029516	1.861551e-02	0.395898	7.844373e-03	0.810275	0.386451	0.514830	0.074887	7.908898e-03	4.903886e-03	0.082705	0.348597	0.277178	6.497335e-03	0.806881	0.736013	0.329685	1.974619e-03	0.937672	0.002336	0.022674	1.987816e-04	2.247728e-04	0.593963	1.080000e+09	0.671574	1.968810e-01	0.806881	0.329685	0.112340	0.645366	0.459750	0.593266	0.314953	0.027597	0.000000	0.810619	2.085213e-03	0.623879	0.605998	0.841179	0.278778	0.026808	0.565252	1.0	0.033798
75%	0.000000	0.535563	0.589157	0.584105	0.613914	0.613842	0.999095	0.797579	0.809469	0.303585	0.781735	4.145000e+09	3.450000e+09	0.471004	5.325533e-04	0.205841	0.199570	0.199570	0.199612	0.238820	0.328623	4.635722e-02	0.116155	0.193493	0.022153	0.848123	0.689647	0.689647	0.217622	7.390000e+09	4.993621e-04	0.264388	0.386731	1.626953e-02	1.224911e-02	0.631125	9.273293e-03	0.148804	0.927109	0.006847	0.376271	0.005764	0.115927	0.191607	0.404551	0.176912	1.454759e-03	8.972876e-03	4.620000e+09	3.652371e-03	0.042903	3.585477e-02	0.401851	1.502031e-02	0.850383	0.540594	0.689051	0.161073	1.295091e-02	1.280557e-02	0.119523	0.360915	0.277429	1.114677e-02	0.942027	0.738560	0.332322	9.005946e-03	0.944811	0.002492	0.035930	4.525945e-04	4.900000e+09	0.594002	4.510000e+09	0.671587	3.722000e-01	0.942027	0.332322	0.117106	0.663062	0.464236	0.624769	0.317707	0.038375	0.000000	0.826455	5.269777e-03	0.624168	0.613913	0.842357	0.281449	0.026913	0.565725	1.0	0.052838
max	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	9.990000e+09	9.980000e+09	1.000000	9.900000e+08	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	3.020000e+09	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	9.990000e+09	9.330000e+09	1.000000	1.000000	2.750000e+09	9.230000e+09	1.000000	9.940000e+09	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	9.740000e+09	9.730000e+09	9.990000e+09	9.990000e+09	1.000000	8.810000e+09	1.000000	9.570000e+09	1.000000	1.000000	1.000000	1.000000	8.820000e+09	9.650000e+09	1.000000	1.000000	1.000000	9.910000e+09	1.000000	1.000000	1.000000	9.540000e+09	1.000000	1.000000	1.000000	1.000000e+10	1.000000e+10	1.000000	1.000000e+10	1.000000	8.320000e+09	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	9.820000e+09	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.0	1.000000

	ROA(C) before interest and depreciation before interest	ROA(A) before interest and % after tax	ROA(B) before interest and depreciation after tax	Operating Gross Margin	Realized Sales Gross Margin	Operating Profit Rate	Pre-tax net Interest Rate	After-tax net Interest Rate	Non-industry income and expenditure/revenue	Continuous interest rate (after tax)	Operating Expense Rate	Research and development expense rate	Cash flow rate	Interest-bearing debt interest rate	Tax rate (A)	Net Value Per Share (B)	Net Value Per Share (A)	Net Value Per Share (C)	Persistent EPS in the Last Four Seasons	Cash Flow Per Share	Revenue Per Share (Yuan ¥)	Operating Profit Per Share (Yuan ¥)	Per Share Net profit before tax (Yuan ¥)	Realized Sales Gross Profit Growth Rate	Operating Profit Growth Rate	After-tax Net Profit Growth Rate	Regular Net Profit Growth Rate	Continuous Net Profit Growth Rate	Total Asset Growth Rate	Net Value Growth Rate	Total Asset Return Growth Rate Ratio	Cash Reinvestment %	Current Ratio	Quick Ratio	Interest Expense Ratio	Total debt/Total net worth	Debt ratio %	Net worth/Assets	Long-term fund suitability ratio (A)	Borrowing dependency	Contingent liabilities/Net worth	Operating profit/Paid-in capital	Net profit before tax/Paid-in capital	Inventory and accounts receivable/Net value	Total Asset Turnover	Accounts Receivable Turnover	Average Collection Days	Inventory Turnover Rate (times)	Fixed Assets Turnover Frequency	Net Worth Turnover Rate (times)	Revenue per person	Operating profit per person	Allocation rate per person	Working Capital to Total Assets	Quick Assets/Total Assets	Current Assets/Total Assets	Cash/Total Assets	Quick Assets/Current Liability	Cash/Current Liability	Current Liability to Assets	Operating Funds to Liability	Inventory/Working Capital	Inventory/Current Liability	Current Liabilities/Liability	Working Capital/Equity	Current Liabilities/Equity	Long-term Liability to Current Assets	Retained Earnings to Total Assets	Total income/Total expense	Total expense/Assets	Current Asset Turnover Rate	Quick Asset Turnover Rate	Working capitcal Turnover Rate	Cash Turnover Rate	Cash Flow to Sales	Fixed Assets to Assets	Current Liability to Liability	Current Liability to Equity	Equity to Long-term Liability	Cash Flow to Total Assets	Cash Flow to Liability	CFO to Assets	Cash Flow to Equity	Current Liability to Current Assets	Liability-Assets Flag	Net Income to Total Assets	Total assets to GNP price	No-credit Interval	Gross Profit to Sales	Net Income to Stockholder's Equity	Liability to Equity	Degree of Financial Leverage (DFL)	Interest Coverage Ratio (Interest expense to EBIT)	Equity to Liability
0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000
0.010000	0.337876	0.352479	0.371831	0.580794	0.581259	0.997582	0.795218	0.807315	0.302198	0.779568	0.000102	0.000000	0.437551	0.000000	0.000000	0.140514	0.140028	0.140028	0.163861	0.281974	0.001470	0.066636	0.122775	0.021671	0.846386	0.678493	0.678509	0.216612	0.000102	0.000324	0.261876	0.321131	0.002390	0.000408	0.624236	0.000493	0.014149	0.761333	0.004911	0.369637	0.005366	0.067003	0.127517	0.393922	0.008996	0.000335	0.000410	0.000102	0.000104	0.010352	0.002279	0.357285	0.000226	0.684405	0.040822	0.085777	0.002648	0.000549	0.000173	0.011704	0.284612	0.273258	0.000000	0.205660	0.723198	0.326493	0.000000	0.853982	0.001835	0.004293	0.000102	0.000102	0.593794	0.000103	0.671375	0.003751	0.205660	0.326493	0.110933	0.523566	0.394009	0.429163	0.295183	0.003790	0.000000	0.679775	0.000174	0.615426	0.580790	0.826098	0.275102	0.025322	0.555236	0.013800
0.100000	0.442344	0.491529	0.491782	0.596513	0.596542	0.998865	0.797171	0.809113	0.303331	0.781369	0.000121	0.000000	0.457875	0.000000	0.000000	0.162606	0.162606	0.162606	0.201097	0.311594	0.009284	0.088674	0.158003	0.022016	0.847886	0.688892	0.688895	0.217539	0.000185	0.000413	0.263353	0.365020	0.005750	0.002625	0.630170	0.001619	0.044055	0.816066	0.005071	0.369637	0.005366	0.088804	0.157649	0.395706	0.047976	0.000544	0.002646	0.000123	0.000145	0.016774	0.006524	0.387212	0.001569	0.748224	0.144206	0.238380	0.014580	0.003140	0.000793	0.034457	0.332352	0.276971	0.001048	0.460519	0.731773	0.327296	0.000000	0.916344	0.002077	0.009190	0.000119	0.000116	0.593912	0.000141	0.671546	0.033545	0.460519	0.327296	0.110933	0.609768	0.450234	0.530051	0.309412	0.010763	0.000000	0.769568	0.000479	0.622950	0.596514	0.837845	0.275939	0.026650	0.564583	0.019089
0.250000	0.476527	0.535543	0.527277	0.600445	0.600434	0.998969	0.797386	0.809312	0.303466	0.781567	0.000157	0.000128	0.461558	0.000203	0.000000	0.173613	0.173613	0.173676	0.214711	0.317748	0.015631	0.096083	0.170370	0.022065	0.847984	0.689270	0.689270	0.217580	4860000000.000000	0.000441	0.263759	0.374749	0.007555	0.004726	0.630612	0.003007	0.072891	0.851196	0.005244	0.370168	0.005366	0.096105	0.169376	0.397403	0.076462	0.000710	0.004387	0.000173	0.000233	0.021774	0.010433	0.392438	0.004121	0.774309	0.241973	0.352845	0.033543	0.005240	0.001973	0.053301	0.341023	0.277034	0.003163	0.626981	0.733612	0.328096	0.000000	0.931097	0.002236	0.014567	0.000146	0.000142	0.593934	0.000274	0.671565	0.085360	0.626981	0.328096	0.110933	0.633265	0.457116	0.565987	0.312995	0.018034	0.000000	0.796750	0.000904	0.623636	0.600443	0.840115	0.276944	0.026791	0.565158	0.024477
0.500000	0.502706	0.559802	0.552278	0.605997	0.605976	0.999022	0.797464	0.809375	0.303525	0.781635	0.000278	509000000.000000	0.465080	0.000321	0.073489	0.184400	0.184400	0.184400	0.224544	0.322487	0.027376	0.104226	0.179709	0.022102	0.848044	0.689439	0.689439	0.217598	6400000000.000000	0.000462	0.264050	0.380425	0.010587	0.007412	0.630698	0.005546	0.111407	0.888593	0.005665	0.372624	0.005366	0.104133	0.178456	0.400131	0.118441	0.000968	0.006573	0.000765	0.000593	0.029516	0.018616	0.395898	0.007844	0.810275	0.386451	0.514830	0.074887	0.007909	0.004904	0.082705	0.348597	0.277178	0.006497	0.806881	0.736013	0.329685	0.001975	0.937672	0.002336	0.022674	0.000199	0.000225	0.593963	1080000000.000000	0.671574	0.196881	0.806881	0.329685	0.112340	0.645366	0.459750	0.593266	0.314953	0.027597	0.000000	0.810619	0.002085	0.623879	0.605998	0.841179	0.278778	0.026808	0.565252	0.033798
0.750000	0.535563	0.589157	0.584105	0.613914	0.613842	0.999095	0.797579	0.809469	0.303585	0.781735	4145000000.000000	3450000000.000000	0.471004	0.000533	0.205841	0.199570	0.199570	0.199612	0.238820	0.328623	0.046357	0.116155	0.193493	0.022153	0.848123	0.689647	0.689647	0.217622	7390000000.000000	0.000499	0.264388	0.386731	0.016270	0.012249	0.631125	0.009273	0.148804	0.927109	0.006847	0.376271	0.005764	0.115927	0.191607	0.404551	0.176912	0.001455	0.008973	4620000000.000000	0.003652	0.042903	0.035855	0.401851	0.015020	0.850383	0.540594	0.689051	0.161073	0.012951	0.012806	0.119523	0.360915	0.277429	0.011147	0.942027	0.738560	0.332322	0.009006	0.944811	0.002492	0.035930	0.000453	4900000000.000000	0.594002	4510000000.000000	0.671587	0.372200	0.942027	0.332322	0.117106	0.663062	0.464236	0.624769	0.317707	0.038375	0.000000	0.826455	0.005270	0.624168	0.613913	0.842357	0.281449	0.026913	0.565725	0.052838
0.900000	0.573948	0.626428	0.620590	0.623153	0.623088	0.999192	0.797713	0.809588	0.303720	0.781856	7980000000.000000	6160000000.000000	0.480048	0.000785	0.269967	0.224828	0.224904	0.224938	0.258788	0.337023	0.076958	0.133947	0.213452	0.022266	0.848285	0.690014	0.690009	0.217663	8390000000.000000	0.000583	0.264969	0.394263	0.027156	0.021649	0.632298	0.014992	0.183934	0.955945	0.010276	0.380774	0.006660	0.133750	0.210627	0.411261	0.254873	0.002480	0.011756	7990000000.000000	5600000000.000000	0.065000	0.079132	0.414138	0.031668	0.892001	0.684313	0.827327	0.299586	0.022673	0.030686	0.159749	0.381650	0.277953	0.018629	0.990066	0.740729	0.336635	0.020790	0.952976	0.002708	0.054500	7081999999.999998	8230000000.000000	0.594067	7381999999.999998	0.671613	0.558893	0.990066	0.336635	0.123092	0.697145	0.475959	0.658502	0.322774	0.049892	0.000000	0.844854	0.013603	0.624699	0.623152	0.843529	0.285573	0.027278	0.566989	0.086886
0.990000	0.664382	0.720662	0.708394	0.652017	0.651990	0.999403	0.798157	0.809967	0.304820	0.782268	9790000000.000000	9440000000.000000	0.516604	758199999.999997	0.643561	0.310391	0.310391	0.310391	0.332121	0.368013	0.215246	0.200606	0.282760	0.024486	0.850907	0.697283	0.696784	0.218559	9758199999.999996	0.001199	0.267758	0.421013	0.074595	0.065860	0.638289	0.039333	0.238667	0.985851	0.048392	0.400201	0.010043	0.199340	0.276928	0.434200	0.522969	0.025452	0.022410	9730000000.000000	9460000000.000000	0.175442	0.277771	0.505090	0.165605	0.959736	0.866496	0.958149	0.676546	0.066620	0.215212	0.225941	0.467819	0.283654	1415599999.999977	1.000000	0.743964	0.354053	1398199999.999997	0.970548	0.003721	0.128899	9788199999.999996	9830000000.000000	0.594542	9620000000.000000	0.671775	0.790928	1.000000	0.354053	0.141850	0.806048	0.552166	0.735353	0.338218	0.113040	0.000000	0.888286	0.166552	0.633377	0.652014	0.846425	0.300099	0.036506	0.572975	0.233943
1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	9990000000.000000	9980000000.000000	1.000000	990000000.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	3020000000.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	9990000000.000000	9330000000.000000	1.000000	1.000000	2750000000.000000	9230000000.000000	1.000000	9940000000.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	9740000000.000000	9730000000.000000	9990000000.000000	9990000000.000000	1.000000	8810000000.000000	1.000000	9570000000.000000	1.000000	1.000000	1.000000	1.000000	8820000000.000000	9650000000.000000	1.000000	1.000000	1.000000	9910000000.000000	1.000000	1.000000	1.000000	9540000000.000000	1.000000	1.000000	1.000000	10000000000.000000	10000000000.000000	1.000000	10000000000.000000	1.000000	8320000000.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	9820000000.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000

	Bankrupt?	ROA(C) before interest and depreciation before interest	Operating Gross Margin	Operating Profit Rate	Non-industry income and expenditure/revenue	Operating Expense Rate	Research and development expense rate	Cash flow rate	Interest-bearing debt interest rate	Net Value Per Share (B)	Persistent EPS in the Last Four Seasons	Cash Flow Per Share	Revenue Per Share (Yuan ¥)	Realized Sales Gross Profit Growth Rate	Operating Profit Growth Rate	After-tax Net Profit Growth Rate	Continuous Net Profit Growth Rate	Total Asset Growth Rate	Net Value Growth Rate	Total Asset Return Growth Rate Ratio	Cash Reinvestment %	Current Ratio	Quick Ratio	Interest Expense Ratio	Total debt/Total net worth	Debt ratio %	Net worth/Assets	Long-term fund suitability ratio (A)	Borrowing dependency	Contingent liabilities/Net worth	Inventory and accounts receivable/Net value	Total Asset Turnover	Accounts Receivable Turnover	Average Collection Days	Inventory Turnover Rate (times)	Fixed Assets Turnover Frequency	Net Worth Turnover Rate (times)	Revenue per person	Operating profit per person	Allocation rate per person	Working Capital to Total Assets	Quick Assets/Total Assets	Current Assets/Total Assets	Cash/Total Assets	Quick Assets/Current Liability	Cash/Current Liability	Current Liability to Assets	Inventory/Working Capital	Inventory/Current Liability	Current Liabilities/Liability	Working Capital/Equity	Long-term Liability to Current Assets	Retained Earnings to Total Assets	Total income/Total expense	Total expense/Assets	Current Asset Turnover Rate	Quick Asset Turnover Rate	Working capitcal Turnover Rate	Cash Turnover Rate	Fixed Assets to Assets	Equity to Long-term Liability	Cash Flow to Total Assets	Cash Flow to Liability	CFO to Assets	Cash Flow to Equity	Current Liability to Current Assets	Total assets to GNP price	No-credit Interval	Net Income to Stockholder's Equity	Degree of Financial Leverage (DFL)	Interest Coverage Ratio (Interest expense to EBIT)	Equity to Liability
0	1	0.370594	0.601457	0.998969	0.302646	1.256969e-04	0.0	0.458143	0.000725	0.147950	0.169141	0.311664	0.017560	0.022102	0.848195	0.688979	0.217535	4.980000e+09	0.000327	0.263100	0.363725	0.002259	0.001208	0.629951	0.021266	0.207576	0.792424	0.005024	0.390284	0.006479	0.398036	0.086957	0.001814	0.003487	1.820926e-04	1.165007e-04	0.032903	0.034164	0.392913	0.037135	0.672775	0.166673	0.190643	0.004094	0.001997	1.473360e-04	0.147308	0.276920	0.001036	0.676269	0.721275	0.025592	0.903225	0.002022	0.064856	7.010000e+08	6.550000e+09	0.593831	4.580000e+08	0.424206	0.126549	0.637555	0.458609	0.520382	0.312905	0.118250	0.009219	0.622879	0.827890	0.026601	0.564050	0.016469
1	1	0.464291	0.610235	0.998946	0.303556	2.897851e-04	0.0	0.461867	0.000647	0.182251	0.208944	0.318137	0.021144	0.022080	0.848088	0.689693	0.217620	6.110000e+09	0.000443	0.264516	0.376709	0.006016	0.004039	0.635172	0.012502	0.171176	0.828824	0.005059	0.376760	0.005835	0.397725	0.064468	0.001286	0.004917	9.360000e+09	7.190000e+08	0.025484	0.006889	0.391590	0.012335	0.751111	0.127236	0.182419	0.014948	0.004136	1.383910e-03	0.056963	0.289642	0.005210	0.308589	0.731975	0.023947	0.931065	0.002226	0.025516	1.065198e-04	7.700000e+09	0.593916	2.490000e+09	0.468828	0.120916	0.641100	0.459001	0.567101	0.314163	0.047775	0.008323	0.623652	0.839969	0.264577	0.570175	0.020794
2	1	0.426071	0.601450	0.998857	0.302035	2.361297e-04	25500000.0	0.458521	0.000790	0.177911	0.180581	0.307102	0.005944	0.022760	0.848094	0.689463	0.217601	7.280000e+09	0.000396	0.264184	0.368913	0.011543	0.005348	0.629631	0.021248	0.207516	0.792484	0.005100	0.379093	0.006562	0.406580	0.014993	0.001495	0.004227	6.500000e+07	2.650000e+09	0.013387	0.028997	0.381968	0.141016	0.829502	0.340201	0.602806	0.000991	0.006302	5.340000e+09	0.098162	0.277456	0.013879	0.446027	0.742729	0.003715	0.909903	0.002060	0.021387	1.791094e-03	1.022676e-03	0.594502	7.610000e+08	0.276179	0.117922	0.642765	0.459254	0.538491	0.314515	0.025346	0.040003	0.623841	0.836774	0.026555	0.563706	0.016474
3	1	0.399844	0.583541	0.998700	0.303350	1.078888e-04	0.0	0.465705	0.000449	0.154187	0.193722	0.321674	0.014368	0.022046	0.848005	0.689110	0.217568	4.880000e+09	0.000382	0.263371	0.384077	0.004194	0.002896	0.630228	0.009572	0.151465	0.848535	0.005047	0.379743	0.005366	0.397925	0.089955	0.001966	0.003215	7.130000e+09	9.150000e+09	0.028065	0.015463	0.378497	0.021320	0.725754	0.161575	0.225815	0.018851	0.002961	1.010646e-03	0.098715	0.276580	0.003540	0.615848	0.729825	0.022165	0.906902	0.001831	0.024161	8.140000e+09	6.050000e+09	0.593889	2.030000e+09	0.559144	0.120760	0.579039	0.448518	0.604105	0.302382	0.067250	0.003252	0.622929	0.834697	0.026697	0.564663	0.023982
4	1	0.465022	0.598783	0.998973	0.303475	7.890000e+09	0.0	0.462746	0.000686	0.167502	0.212537	0.319162	0.029690	0.022096	0.848258	0.689697	0.217626	5.510000e+09	0.000439	0.265218	0.379690	0.006022	0.003727	0.636055	0.005150	0.106509	0.893491	0.005303	0.375025	0.006624	0.400079	0.175412	0.001449	0.004367	1.633674e-04	2.935211e-04	0.040161	0.058111	0.394371	0.023988	0.751822	0.260330	0.358380	0.014161	0.004275	6.804636e-04	0.110195	0.287913	0.004869	0.975007	0.732000	0.000000	0.913850	0.002224	0.026385	6.680000e+09	5.050000e+09	0.593915	8.240000e+08	0.309555	0.110933	0.622374	0.454411	0.578469	0.311567	0.047725	0.003878	0.623521	0.839973	0.024752	0.575617	0.035490

	0
Operating Profit Growth Rate	-71.688950
Operating Profit Rate	-70.237164
Net Income to Stockholder's Equity	-37.964701
Working Capital/Equity	-36.203654
Working capitcal Turnover Rate	-28.584611

	0
Contingent liabilities/Net worth	79.670620
Net Value Growth Rate	80.291844
Total income/Total expense	82.332424
Current Ratio	82.577237
Fixed Assets to Assets	82.577237

	Realized Sales Gross Profit Growth Rate	Contingent liabilities/Net worth	Net Value Growth Rate	Total income/Total expense	Current Ratio	Fixed Assets to Assets	Bankrupt?
0	0.022121	0.005366	0.000445	0.002584	0.058422	0.983363	0
1	0.022092	0.005366	0.000784	0.001958	0.003541	0.957850	0
2	0.022124	0.010482	0.000474	0.005955	2750000000.000000	0.057384	0
3	0.022131	0.005366	0.000445	0.002232	0.000555	0.914384	0
4	0.022061	0.005366	0.000464	0.002236	0.001725	0.968198	0
5	0.021967	0.005366	0.000400	0.001869	0.000193	0.918705	0
6	0.022216	0.005366	1350000000.000000	0.001892	0.000000	0.882981	0
7	1.000000	0.005366	0.001381	0.002231	0.009978	0.032239	0
8	0.022096	0.005366	0.000444	0.002500	0.055196	0.983645	0
9	0.022155	0.005366	0.000534	0.002266	0.000566	0.901367	0
10	0.022052	0.006517	0.000454	1.000000	0.000604	0.002138	0
11	0.021601	0.006179	0.000520	0.000000	1.000000	0.000000	0
12	0.021992	0.005366	0.000443	0.002184	0.003734	0.998725	0
13	0.022073	0.005366	0.000444	0.002444	0.031061	0.941900	0
14	0.022092	0.005848	1.000000	0.002157	0.007650	0.168345	0
15	0.021528	1.000000	0.000204	0.001647	0.001086	0.620196	1
16	0.022107	0.005366	0.000437	0.002121	0.008113	8320000000.000000	1
17	0.021732	0.004445	9330000000.000000	0.001928	0.001261	0.652032	1
18	0.022193	0.005366	0.000331	0.001891	0.002554	1.000000	1

	Total Asset Return Growth Rate Ratio	Continuous Net Profit Growth Rate
0.000000	0.251620	0.000000
0.010000	0.261903	0.216645
0.100000	0.263354	0.217539
0.250000	0.263760	0.217580
0.500000	0.264050	0.217598
0.750000	0.264389	0.217622
0.900000	0.264969	0.217663
0.990000	0.267758	0.218561
1.000000	1.000000	1.000000

	Inventory/Working Capital	Degree of Financial Leverage (DFL)	Total debt/Total net worth	Quick Assets/Current Liability	Revenue per person	Bankrupt?
0	0.277287	0.026790	821000000.000000	0.222624	0.013431	0
1	0.277230	1.000000	0.004319	0.007689	0.025769	0
2	0.277295	0.026790	5930000000.000000	0.100846	0.066554	0
3	0.276990	0.026791	6470000000.000000	0.148478	0.049584	0
4	0.277251	0.026772	1190000000.000000	0.266294	0.000632	0
5	0.277719	0.026802	0.013172	0.005739	1.000000	0
6	0.277005	0.026791	1820000000.000000	0.251210	0.002858	0
7	0.278049	0.026793	0.008267	8140000000.000000	0.158101	0
8	0.277302	0.026875	0.011483	0.006766	0.991117	0
9	0.277289	0.026793	9940000000.000000	0.198357	0.074189	0
10	0.277333	0.026851	0.007450	0.006928	0.840788	0
11	1.000000	0.026968	0.002708	0.002510	0.006422	0
12	0.279411	0.027092	0.018772	8820000000.000000	0.107607	0
13	0.276975	0.026620	1.000000	0.001226	0.093019	1
14	0.276392	0.026351	0.044451	0.000253	7050000000.000000	1
15	0.276975	0.026764	3470000000.000000	0.107927	0.036050	1

	Average Collection Days	Quick Ratio	Equity to Long-term Liability	Non-industry income and expenditure/revenue	Revenue Per Share (Yuan ¥)	Bankrupt?
0	1790000000.000000	0.000283	0.115993	0.303503	0.015049	0
1	0.000582	5800000000.000000	0.110933	0.303387	0.008848	0
2	2130000000.000000	0.000656	0.110933	0.303293	0.001981	0
3	8800000000.000000	0.001520	0.110933	0.303550	0.160655	0
4	273000000.000000	0.042353	0.110933	0.303593	0.009135	0
5	0.001020	0.003398	0.110933	0.303526	0.993285	0
6	0.016404	0.005301	0.114850	1.000000	0.000000	0
7	751000000.000000	0.019313	0.119212	0.303552	0.040232	0
8	0.002247	0.005004	0.110933	0.303526	1.000000	0
9	478000000.000000	0.004243	0.110933	0.303565	0.036935	0
10	0.000202	8920000000.000000	0.118868	0.303523	0.031293	0
11	7940000000.000000	0.000248	0.110933	0.303544	0.022369	0
12	0.000213	8480000000.000000	0.115753	0.303519	0.015019	0
13	0.000211	4800000000.000000	0.110933	0.303649	0.003131	0
14	1790000000.000000	0.001490	0.111470	0.320065	0.001270	0
15	0.065120	0.006201	0.111182	0.730252	1510000000.000000	0
16	0.001289	3490000000.000000	0.110933	0.303413	0.011737	0
17	3240000000.000000	0.000659	0.112014	0.303541	0.030960	0
18	0.003621	8170000000.000000	0.110933	0.300096	0.002012	0
19	598000000.000000	0.002935	0.114540	0.330080	0.001724	0
20	4350000000.000000	0.000244	0.110933	0.303493	0.027103	0
21	0.000577	0.322045	0.110933	0.297521	3020000000.000000	0
22	5810000000.000000	0.002072	0.112525	0.303547	0.038659	0
23	8370000000.000000	0.004860	0.110933	0.303554	0.040171	0
24	0.000541	5240000000.000000	0.110933	0.303503	0.036814	0
25	2480000000.000000	2990000000.000000	0.112071	0.303399	0.024154	0
26	2480000000.000000	2990000000.000000	0.112071	0.303399	0.024154	0
27	0.131772	0.001199	0.112778	0.446321	1510000000.000000	0
28	7860000000.000000	0.006972	0.111496	0.303511	0.029584	0
29	0.005974	0.004721	1.000000	0.303529	0.145106	0
30	478000000.000000	0.000778	0.120176	0.303529	0.039309	0
31	137000000.000000	0.000000	0.110933	0.303296	0.003796	1
32	0.014798	9230000000.000000	0.116022	0.235090	0.000136	1
33	0.000677	0.000996	0.862994	0.303528	0.166039	1
34	0.010649	0.001714	0.922128	0.302381	0.018740	1

	Average Collection Days	Quick Ratio	Equity to Long-term Liability	Non-industry income and expenditure/revenue	Revenue Per Share (Yuan ¥)
0.000000	0.000000	0.000100	0.000000	0.271546	0.000106
0.010000	0.000442	0.000442	0.110933	0.302337	0.001800
0.100000	0.002694	0.002678	0.110933	0.303332	0.009423
0.250000	0.004401	0.004752	0.110933	0.303466	0.015715
0.500000	0.006577	0.007434	0.112361	0.303525	0.027406
0.750000	0.008955	0.012213	0.117117	0.303585	0.046365
0.900000	0.011703	0.021503	0.123095	0.303718	0.076694
0.990000	0.019747	0.061341	0.141209	0.304698	0.205835
1.000000	0.718671	0.272800	0.480440	0.321553	0.720964

	Borrowing dependency	Total assets to GNP price	Long-term fund suitability ratio (A)	Accounts Receivable Turnover	Allocation rate per person	Bankrupt?
0	0.373670	0.000782	0.029761	0.000797	8520000000.000000	0
1	0.416696	0.020016	0.005634	304000000.000000	0.043170	0
2	0.369637	8780000000.000000	0.009171	0.000657	0.001214	0
3	0.400874	0.018425	0.005599	3650000000.000000	0.043680	0
4	0.369637	7420000000.000000	0.008207	0.009962	0.000723	0
5	0.382681	9820000000.000000	0.006637	0.000864	0.000850	0
6	0.369637	7790000000.000000	0.005471	0.003143	0.003505	0
7	0.406525	0.020289	0.005886	3250000000.000000	0.034783	0
8	0.371015	0.000486	0.047691	0.001057	9570000000.000000	0
9	0.382589	5260000000.000000	0.005207	0.001079	0.000521	0
10	0.402899	0.008384	0.004842	7300000000.000000	0.413331	0
11	0.402818	0.007000	0.004852	4460000000.000000	0.397763	0
12	0.369758	8880000000.000000	0.006392	9330000000.000000	0.001231	0
13	0.369758	8880000000.000000	0.006392	9330000000.000000	0.001231	0
14	0.369637	9390000000.000000	0.005811	0.001045	0.004742	0
15	0.403770	0.003720	0.006372	1420000000.000000	0.039502	0
16	0.383563	6410000000.000000	0.004993	0.000580	0.003697	0
17	0.371300	0.000666	0.040531	0.000854	6210000000.000000	0
18	0.369770	0.000331	0.069853	0.001349	8280000000.000000	0
19	0.369637	0.047880	0.005028	0.003332	1.000000	0
20	0.380369	9650000000.000000	0.008541	0.004419	0.001473	0
21	0.369637	1920000000.000000	0.005577	0.002041	0.001432	0
22	0.437324	0.023522	0.005659	2840000000.000000	0.033298	0
23	0.403900	0.019353	0.005695	3250000000.000000	0.039626	0
24	0.373378	4880000000.000000	0.005331	0.000521	0.007362	0
25	0.369637	8890000000.000000	0.013708	0.009741	0.000349	0
26	0.404433	0.019708	0.005648	3650000000.000000	0.045817	0
27	0.395236	4370000000.000000	0.006144	0.021793	0.000900	0
28	0.357440	3030000000.000000	0.000000	0.001142	0.001259	0
29	0.369715	0.003876	0.011633	994000000.000000	0.005058	0
30	0.373412	0.001765	0.005234	7910000000.000000	0.153342	0
31	0.392974	0.006550	0.004884	4870000000.000000	0.418038	0
32	0.372707	0.000745	0.045113	0.000820	5290000000.000000	0
33	0.371337	394000000.000000	0.005541	0.000970	0.002772	0
34	0.372278	0.010129	0.007632	7300000000.000000	0.148772	0
35	0.401300	0.005256	0.004945	304000000.000000	0.212764	0
36	0.374354	0.000485	0.060427	0.001128	8410000000.000000	0
37	0.369897	0.001326	0.145681	0.000623	7970000000.000000	0
38	0.374084	0.002995	0.254191	0.000359	23200000.000000	0
39	0.385721	6860000000.000000	0.004952	0.000392	0.003878	0
40	0.385257	0.018980	1.000000	0.013612	0.000185	0
41	0.369637	0.002582	0.608185	0.001114	8900000000.000000	0
42	0.465365	0.002302	0.005076	812000000.000000	0.051331	0
43	0.370128	0.001094	0.085521	0.000629	2190000000.000000	0
44	0.372771	0.013550	0.984855	0.001108	7910000000.000000	0
45	0.372771	0.013550	0.984855	0.001108	7910000000.000000	0
46	0.369637	2710000000.000000	0.005346	0.000781	0.003710	1
47	0.431840	8140000000.000000	0.005328	0.000684	0.001540	1
48	0.954819	0.003198	0.004826	0.009341	0.037792	1
49	0.395616	0.002044	0.923930	0.450399	3480000000.000000	1
50	0.395616	0.002044	0.923930	0.450399	3480000000.000000	1
51	1.000000	0.000456	0.004902	0.000594	0.009052	1
52	0.369637	9170000000.000000	0.006451	0.001057	0.000667	1
53	0.409987	0.003736	0.004874	1220000000.000000	0.013696	1

	Inventory and accounts receivable/Net value	Current Liability to Current Assets	Cash/Current Liability	Cash Flow to Equity
0.000000	0.000000	0.000770	0.000000	0.000000
0.010000	0.393995	0.004009	0.000174	0.295908
0.100000	0.395740	0.010881	0.000817	0.309430
0.250000	0.397424	0.018009	0.001987	0.313004
0.500000	0.400123	0.027500	0.004925	0.314956
0.750000	0.404469	0.038225	0.012796	0.317707
0.900000	0.410932	0.049704	0.030290	0.322786
0.990000	0.431420	0.109633	0.177336	0.338199
1.000000	0.707445	0.650661	9650000000.000000	1.000000

	Inventory and accounts receivable/Net value	Current Liability to Current Assets	Cash/Current Liability	Cash Flow to Equity	Bankrupt?
0	0.426719	0.034957	7580000000.000000	0.313553	0
1	0.414449	0.033030	8550000000.000000	0.312888	0
2	0.394885	0.040896	9060000000.000000	0.310667	0
3	0.399834	0.043580	5900000000.000000	0.314501	0
4	0.428287	0.033204	9170000000.000000	0.314467	0
5	0.419155	0.029034	1300000000.000000	0.314287	0
6	0.407095	0.037160	4610000000.000000	0.311171	0
7	0.425422	0.033021	4610000000.000000	0.312245	0
8	0.398329	0.042739	9650000000.000000	0.313738	0
9	0.415081	0.059007	1840000000.000000	0.310570	0
10	0.397457	0.112154	6610000000.000000	0.314369	0
11	0.396652	0.075770	8590000000.000000	0.313749	0
12	0.400667	0.050842	8330000000.000000	0.314293	0
13	0.410790	0.035499	7510000000.000000	0.313670	0
14	0.423331	0.040878	8870000000.000000	0.313287	0
15	0.399777	0.055387	6870000000.000000	0.313302	0
16	0.397619	0.045314	3830000000.000000	0.314393	0
17	0.405171	0.033506	9410000000.000000	0.314091	0
18	0.400018	0.044585	8870000000.000000	0.308577	0
19	0.410212	0.041792	287000000.000000	0.314553	0
20	0.413891	0.030715	6970000000.000000	0.314694	0
21	0.430161	0.031565	5740000000.000000	0.314544	0
22	0.406061	0.089706	7540000000.000000	0.313538	0
23	0.406961	0.070746	0.012010	1.000000	0
24	0.419876	0.065144	3080000000.000000	0.313816	0
25	0.396269	0.080904	4950000000.000000	0.314529	0
26	0.406580	0.025346	5340000000.000000	0.314515	1
27	0.425839	0.042234	3370000000.000000	0.309981	1
28	0.408737	0.080395	444000000.000000	0.286312	1
29	0.441273	0.035799	2090000000.000000	0.312128	1
30	0.413586	0.084561	3110000000.000000	0.314222	1
31	0.396232	0.142964	6730000000.000000	0.314092	1
32	0.411399	0.055279	696000000.000000	0.314013	1
33	0.415298	0.055361	5390000000.000000	0.311226	1
34	0.406098	0.069956	9010000000.000000	0.310119	1
35	0.434070	0.104796	1140000000.000000	0.306167	1
36	0.415443	0.037514	3540000000.000000	0.313609	1
37	0.397559	0.084685	248000000.000000	0.306692	1
38	0.400885	0.023458	479000000.000000	0.313628	1
39	0.416035	0.021729	6840000000.000000	0.314395	1
40	0.405900	0.042721	7660000000.000000	0.314273	1

	Interest-bearing debt interest rate	Equity to Liability	Operating profit per person	Cash Flow Per Share	Net Worth Turnover Rate (times)	Total expense/Assets
0.000000	0.000000	0.003946	0.000000	0.128740	0.009032	0.000853
0.010000	0.000000	0.014116	0.359405	0.282839	0.010806	0.004505
0.100000	0.000000	0.019273	0.387479	0.311771	0.017097	0.009379
0.250000	0.000205	0.024618	0.392472	0.317818	0.021935	0.014746
0.500000	0.000321	0.033961	0.395920	0.322558	0.029677	0.022767
0.750000	0.000534	0.052997	0.401777	0.328712	0.042903	0.035914
0.900000	0.000786	0.086714	0.413756	0.337094	0.064548	0.054298
0.990000	750000000.000000	0.223302	0.495997	0.367599	0.171310	0.121626
1.000000	990000000.000000	0.798122	1.000000	0.577386	0.510645	1.000000

	Interest-bearing debt interest rate	Equity to Liability	Operating profit per person	Cash Flow Per Share	Net Worth Turnover Rate (times)	Total expense/Assets
0	520000000.000000	0.072854	0.396418	0.325387	0.019032	0.027888
1	490000000.000000	0.031042	0.410563	0.333805	0.032581	0.012962
2	750000000.000000	0.031186	0.407623	0.325564	0.042258	0.022009
3	60000000.000000	0.021747	0.401076	0.306925	0.065645	0.021840
4	70000000.000000	0.042678	0.404819	0.324680	0.048871	0.028932
5	730000000.000000	0.038066	0.401269	0.319375	0.028548	0.018908
6	690000000.000000	0.028336	0.400330	0.338367	0.041290	0.017965
7	440000000.000000	0.089848	0.396181	0.323973	0.034194	0.008222
8	820000000.000000	0.044549	0.432091	0.335432	0.035323	0.007565
9	620000000.000000	0.024016	0.418749	0.325953	0.028548	0.008602
10	780000000.000000	0.073980	0.397537	0.328146	0.015323	0.012923
11	360000000.000000	0.071542	0.393987	0.387635	0.039194	0.009749
12	8000000.000000	0.064913	0.401144	0.330480	0.022419	0.028847
13	840000000.000000	0.106411	0.426585	0.326943	0.014839	0.005934
14	380000000.000000	0.045433	0.400579	0.334335	0.027258	0.015545
15	350000000.000000	0.027587	0.393716	0.324786	0.039194	0.022653
16	3000000.000000	0.082486	0.379933	0.309153	0.010645	0.066169
17	260000000.000000	0.037803	0.396463	0.323548	0.028065	0.026794
18	320000000.000000	0.040429	0.416872	0.329702	0.019032	0.009266
19	960000000.000000	0.027628	0.402467	0.327509	0.039355	0.016886
20	2000000.000000	0.091038	0.403779	0.334689	0.022581	0.022744
21	660000000.000000	0.073416	0.401574	0.326307	0.023065	0.037820
22	670000000.000000	0.085945	0.396949	0.326943	0.020645	0.027039
23	740000000.000000	0.037684	0.398555	0.321249	0.023387	0.035183
24	610000000.000000	0.028867	0.408856	0.342152	0.048226	0.024128
25	910000000.000000	0.029368	0.403473	0.308234	0.037419	0.038075
26	750000000.000000	0.025821	0.396181	0.323230	0.048871	0.012957
27	2000000.000000	0.026984	0.405192	0.329030	0.039032	0.012321
28	480000000.000000	0.134306	0.395299	0.322098	0.015806	0.013089
29	960000000.000000	0.041323	0.399482	0.314954	0.028226	0.004456
30	520000000.000000	0.036720	0.382952	0.314954	0.029355	0.054901
31	310000000.000000	0.105285	0.404796	0.326484	0.027419	0.026217
32	9000000.000000	0.035611	0.413910	0.342965	0.065000	0.041523
33	550000000.000000	0.020656	0.399403	0.329313	0.094516	0.018354
34	10000000.000000	0.093217	0.410721	0.329278	0.022419	0.016048
35	3000000.000000	0.054352	0.408335	0.339181	0.025161	0.015610
36	8000000.000000	0.050445	0.400669	0.328606	0.023226	0.010499
37	310000000.000000	0.063523	0.407092	0.334088	0.021935	0.032127
38	740000000.000000	0.051965	0.393942	0.325281	0.016290	0.027915
39	460000000.000000	0.062529	0.398872	0.318773	0.024839	0.023611

	Interest-bearing debt interest rate	Equity to Liability	Operating profit per person	Cash Flow Per Share	Net Worth Turnover Rate (times)	Total expense/Assets
0.000000	0.000000	0.003946	0.000000	0.128740	0.009032	0.000895
0.010000	0.000000	0.014077	0.359396	0.282793	0.010806	0.004679
0.100000	0.000000	0.019159	0.387452	0.311771	0.017097	0.009465
0.250000	0.000200	0.024438	0.392449	0.317818	0.021935	0.014799
0.500000	0.000314	0.033580	0.395853	0.322505	0.029839	0.022799
0.750000	0.000499	0.052144	0.401712	0.328606	0.043065	0.035981
0.900000	0.000718	0.085175	0.413582	0.336917	0.064516	0.054118
0.990000	0.002153	0.221338	0.496057	0.367096	0.172250	0.121162
1.000000	0.796080	0.798122	0.927252	0.577386	0.510645	0.368382

	Net Value Per Share (B)	Persistent EPS in the Last Four Seasons
0.000000	0.069656	0.000000
0.010000	0.142050	0.164987
0.100000	0.162863	0.201002
0.250000	0.173613	0.214617
0.500000	0.184147	0.224355
0.750000	0.199191	0.238253
0.900000	0.223720	0.257928
0.990000	0.304567	0.329948
1.000000	0.549197	1.000000

	Operating Gross Margin
0.000000	0.156308
0.010000	0.581603
0.100000	0.596519
0.250000	0.600398
0.500000	0.605897
0.750000	0.613593
0.900000	0.622794
0.990000	0.651947
1.000000	0.665151

	Interest Expense Ratio	Interest Coverage Ratio (Interest expense to EBIT)	No-credit Interval	Retained Earnings to Total Assets
0	0.633188	0.567650	0.000000	0.926604
1	0.606654	0.000000	0.622948	0.940913
2	0.000000	0.707735	0.622525	0.928672

	Interest Expense Ratio	Interest Coverage Ratio (Interest expense to EBIT)	No-credit Interval	Retained Earnings to Total Assets
0.000000	0.459985	0.172065	0.408682	0.594171
0.010000	0.624199	0.554684	0.615353	0.857615
0.100000	0.630141	0.564537	0.622970	0.916733
0.250000	0.630612	0.565158	0.623635	0.931024
0.500000	0.630709	0.565268	0.623874	0.937620
0.750000	0.631151	0.565747	0.624153	0.944626
0.900000	0.632345	0.567040	0.624668	0.952794
0.990000	0.638549	0.572922	0.632929	0.970484
1.000000	1.000000	1.000000	1.000000	1.000000

	Net Income to Stockholder's Equity	Working Capital/Equity	Working capitcal Turnover Rate	After-tax Net Profit Growth Rate	Bankrupt?
0	0.442176	0.000000	0.593870	0.685937	1
1	0.000000	0.517571	0.593862	0.685068	1

	Net Income to Stockholder's Equity	Working Capital/Equity	Working capitcal Turnover Rate	After-tax Net Profit Growth Rate
0.000000	0.344652	0.507149	0.572892	0.180701
0.010000	0.827504	0.723756	0.593802	0.678521
0.100000	0.837850	0.731798	0.593912	0.688887
0.250000	0.840106	0.733609	0.593934	0.689268
0.500000	0.841159	0.735983	0.593962	0.689438
0.750000	0.842323	0.738524	0.594000	0.689648
0.900000	0.843493	0.740703	0.594064	0.690010
0.990000	0.846347	0.743653	0.594440	0.696846
1.000000	1.000000	0.825197	0.605383	1.000000

	Operating Profit Growth Rate	Operating Profit Rate
0.000000	0.736430	0.973424
0.010000	0.846378	0.997955
0.100000	0.847886	0.998870
0.250000	0.847983	0.998970
0.500000	0.848043	0.999022
0.750000	0.848123	0.999093
0.900000	0.848285	0.999190
0.990000	0.851110	0.999399
1.000000	1.000000	0.999778

	columns	score	p_value
0	Retained Earnings to Total Assets	403.773118	0.0000
1	Debt ratio %	382.439927	0.0000
2	Net worth/Assets	382.439927	0.0000
3	ROA(C) before interest and depreciation befor...	372.487581	0.0000
4	Persistent EPS in the Last Four Seasons	310.063059	0.0000
5	Cash/Current Liability	286.252677	0.0000
6	Current Liability to Current Assets	284.595750	0.0000
7	Working Capital to Total Assets	247.313865	0.0000
8	Current Liability to Assets	225.122244	0.0000
9	Borrowing dependency	205.120464	0.0000
10	Net Income to Stockholder's Equity	191.725273	0.0000
11	Net Value Per Share (B)	172.137867	0.0000
12	Liability-Assets Flag	146.025382	0.0000
13	Equity to Long-term Liability	145.974589	0.0000
14	Working Capital/Equity	128.954106	0.0000
15	Non-industry income and expenditure/revenue	110.983253	0.0000
16	Total expense/Assets	90.930228	0.0000
17	Total income/Total expense	85.334556	0.0000
18	Operating Gross Margin	74.137990	0.0000
19	Operating profit per person	64.427347	0.0000
20	Tax rate (A)	62.685816	0.0000
21	Cash/Total Assets	57.975673	0.0000
22	Total assets to GNP price	57.198236	0.0000
23	Quick Assets/Total Assets	52.781185	0.0000
24	Current Ratio	51.061844	0.0000
25	Quick Assets/Current Liability	45.741873	0.0000
26	CFO to Assets	43.589989	0.0000
27	Equity to Liability	42.928794	0.0000
28	Fixed Assets Turnover Frequency	39.380221	0.0000
29	Inventory and accounts receivable/Net value	38.909708	0.0000
30	Operating Profit Rate	38.850426	0.0000
31	Cash flow rate	36.832999	0.0000
32	Total Asset Turnover	34.540194	0.0000
33	Contingent liabilities/Net worth	32.300455	0.0000
34	Average Collection Days	28.537532	0.0000
35	Fixed Assets to Assets	28.537532	0.0000
36	Total debt/Total net worth	28.537532	0.0000
37	Allocation rate per person	28.537532	0.0000
38	Accounts Receivable Turnover	28.537532	0.0000
39	Revenue per person	28.537532	0.0000
40	Net Value Growth Rate	28.537532	0.0000
41	Quick Ratio	28.537532	0.0000
42	Total Asset Return Growth Rate Ratio	28.404717	0.0000
43	Cash Flow Per Share	26.275563	0.0000
44	Working capitcal Turnover Rate	25.605834	0.0000
45	Operating Profit Growth Rate	21.745893	0.0000
46	Cash Flow to Total Assets	20.771819	0.0000
47	Current Assets/Total Assets	19.982075	0.0000
48	Revenue Per Share (Yuan ¥)	18.624912	0.0000
49	Cash Flow to Equity	16.779115	0.0000
50	Total Asset Growth Rate	14.101149	0.0002
51	Cash Reinvestment %	11.383942	0.0007
52	Cash Flow to Liability	10.767865	0.0010
53	After-tax Net Profit Growth Rate	9.833110	0.0017
54	Long-term fund suitability ratio (A)	8.151000	0.0043
55	Research and development expense rate	6.270324	0.0123
56	Cash Turnover Rate	4.142295	0.0419
57	Quick Asset Turnover Rate	3.168890	0.0751
58	Continuous Net Profit Growth Rate	3.054471	0.0806
59	Current Liabilities/Liability	2.525012	0.1121
60	Degree of Financial Leverage (DFL)	1.969514	0.1606
61	Current Asset Turnover Rate	1.738746	0.1874
62	Net Worth Turnover Rate (times)	1.572778	0.2099
63	Realized Sales Gross Profit Growth Rate	1.315570	0.2514
64	Interest Coverage Ratio (Interest expense to ...	0.341999	0.5587
65	Long-term Liability to Current Assets	0.259598	0.6104
66	Inventory/Current Liability	0.187725	0.6648
67	No-credit Interval	0.147786	0.7007
68	Interest-bearing debt interest rate	0.066144	0.7970
69	Operating Expense Rate	0.035123	0.8513
70	Inventory/Working Capital	0.014130	0.9054
71	Interest Expense Ratio	0.010103	0.9199
72	Inventory Turnover Rate (times)	0.005834	0.9391

	Working Capital to Total Assets	Current Liability to Assets	Working Capital/Equity	Tax rate (A)	Average Collection Days	Research and development expense rate	Current Liability to Current Assets	Borrowing dependency	Interest-bearing debt interest rate	Operating Gross Margin	CFO to Assets	Total debt/Total net worth	Equity to Long-term Liability	ROA(C) before interest and depreciation before interest	Total expense/Assets	Total Asset Turnover	Inventory and accounts receivable/Net value	Inventory/Current Liability	Net Value Per Share (B)	Persistent EPS in the Last Four Seasons	Total income/Total expense	Revenue Per Share (Yuan ¥)	Net Income to Stockholder's Equity	Cash/Total Assets	Net Worth Turnover Rate (times)	Equity to Liability	Cash Flow Per Share	Working capitcal Turnover Rate	Cash flow rate	Long-term Liability to Current Assets	Current Ratio	Cash Reinvestment %	Quick Ratio	Retained Earnings to Total Assets	Quick Assets/Current Liability	Accounts Receivable Turnover	Allocation rate per person	Operating Profit Rate	Total Asset Return Growth Rate Ratio	Cash/Current Liability	Net Value Growth Rate	Revenue per person	Total assets to GNP price	Realized Sales Gross Profit Growth Rate	Operating profit per person	Long-term fund suitability ratio (A)	Cash Flow to Equity	Cash Flow to Total Assets	Inventory/Working Capital	Contingent liabilities/Net worth	After-tax Net Profit Growth Rate	Operating Profit Growth Rate	Continuous Net Profit Growth Rate	Non-industry income and expenditure/revenue	No-credit Interval	Cash Flow to Liability	Interest Expense Ratio	Interest Coverage Ratio (Interest expense to EBIT)	Total Asset Growth Rate	Degree of Financial Leverage (DFL)	Current Asset Turnover Rate	Fixed Assets Turnover Frequency	Total
1.5 * IQR	1.180000	1.300000	1.440000	1.740000	2.180000	2.580000	3.460000	3.460000	3.600000	4.620000	4.760000	5.100000	5.260000	5.340000	5.340000	5.420000	5.480000	6.390000	6.710000	6.730000	6.990000	7.070000	7.230000	7.430000	7.510000	7.850000	8.010000	8.030000	8.330000	8.470000	8.650000	8.730000	8.750000	8.790000	9.150000	9.250000	9.370000	9.610000	9.650000	10.490000	10.610000	10.670000	11.550000	11.810000	11.850000	11.930000	12.170000	12.510000	13.230000	13.710000	14.370000	14.390000	14.570000	15.390000	16.390000	17.270000	17.690000	18.800000	19.680000	20.020000	20.360000	20.520000	92.790000
1.7 * IQR	0.460000	0.580000	0.980000	1.500000	1.600000	1.060000	2.820000	2.760000	3.060000	3.840000	3.580000	4.200000	4.260000	3.920000	4.280000	4.580000	4.560000	5.480000	5.580000	5.460000	5.880000	6.290000	6.160000	6.530000	6.690000	6.870000	6.410000	7.030000	6.910000	7.650000	7.730000	7.350000	7.770000	7.070000	8.050000	8.290000	8.390000	7.850000	8.010000	9.650000	9.550000	10.010000	10.710000	10.650000	10.310000	11.270000	10.590000	11.050000	11.810000	12.550000	13.330000	13.050000	13.370000	13.770000	14.650000	15.810000	16.410000	16.970000	19.660000	18.530000	20.140000	20.240000	90.630000
1.8 * IQR	0.240000	0.520000	0.960000	1.360000	1.320000	0.520000	2.560000	2.480000	2.760000	3.540000	3.100000	3.840000	3.760000	3.500000	3.860000	4.120000	4.280000	5.120000	5.200000	4.920000	5.440000	5.820000	5.500000	6.020000	6.240000	6.410000	5.920000	6.570000	6.290000	7.330000	7.370000	6.610000	7.170000	6.370000	7.490000	7.910000	8.010000	7.050000	7.370000	9.270000	8.970000	9.590000	10.230000	10.230000	9.910000	10.870000	9.450000	10.170000	11.350000	12.110000	12.990000	12.470000	12.850000	13.110000	14.090000	15.210000	15.810000	16.150000	18.470000	18.050000	20.000000	20.140000	89.790000
1.9 * IQR	0.140000	0.420000	0.800000	1.260000	1.060000	0.000000	2.380000	2.260000	2.640000	3.320000	2.720000	3.560000	3.440000	2.960000	3.540000	3.780000	3.940000	4.720000	4.800000	4.460000	4.960000	5.540000	4.760000	5.500000	5.800000	6.020000	5.320000	6.140000	5.900000	6.910000	6.990000	6.120000	6.810000	5.760000	7.130000	7.630000	7.530000	6.510000	6.770000	8.930000	8.630000	9.230000	9.830000	9.950000	9.150000	10.590000	8.730000	9.550000	10.910000	11.550000	12.490000	12.010000	12.610000	12.570000	13.530000	14.510000	15.310000	15.550000	13.210000	17.610000	19.880000	20.020000	87.830000
2 * IQR	0.140000	0.240000	0.700000	1.160000	0.980000	0.000000	2.260000	2.120000	2.420000	3.060000	2.360000	3.200000	3.140000	2.640000	3.200000	3.460000	3.680000	4.500000	4.420000	4.240000	4.540000	5.120000	4.400000	5.180000	5.540000	5.700000	4.820000	5.800000	5.700000	6.590000	6.530000	5.620000	6.470000	5.240000	6.770000	7.250000	7.210000	6.100000	6.180000	8.650000	8.230000	8.790000	9.310000	9.630000	8.710000	10.310000	8.150000	8.870000	10.570000	10.970000	12.190000	11.630000	12.210000	11.990000	13.050000	13.890000	14.790000	14.990000	12.270000	17.090000	19.820000	19.960000	86.930000
2.1 * IQR	0.060000	0.120000	0.660000	1.000000	0.900000	0.000000	2.120000	2.000000	2.360000	2.840000	1.980000	3.020000	2.820000	2.160000	3.040000	3.240000	3.240000	4.320000	4.080000	3.840000	4.200000	4.940000	3.980000	4.860000	5.420000	5.280000	4.480000	5.460000	5.420000	6.270000	6.180000	5.360000	6.060000	4.960000	6.390000	6.910000	6.930000	5.500000	5.640000	8.430000	7.890000	8.470000	8.970000	9.270000	8.370000	10.090000	7.610000	8.190000	10.130000	10.370000	11.710000	11.190000	11.890000	11.470000	12.530000	13.210000	14.310000	14.530000	12.230000	16.570000	19.740000	19.860000	86.110000
2.2 * IQR	0.040000	0.060000	0.620000	0.940000	0.740000	0.000000	1.920000	1.740000	2.320000	2.580000	1.640000	2.680000	2.620000	1.900000	2.880000	2.860000	3.000000	4.080000	3.860000	3.620000	4.000000	4.720000	3.620000	4.720000	5.220000	4.940000	4.220000	5.200000	5.040000	5.860000	5.780000	4.880000	5.840000	4.500000	6.080000	6.710000	6.650000	5.200000	5.220000	8.250000	7.490000	8.190000	8.730000	8.870000	7.790000	9.810000	7.170000	7.690000	9.870000	10.130000	11.230000	10.750000	11.530000	11.030000	12.030000	12.870000	13.850000	14.050000	0.000000	16.050000	19.640000	19.760000	84.390000
2.5 * IQR	0.020000	0.000000	0.480000	0.660000	0.560000	0.000000	1.620000	1.360000	2.000000	2.180000	1.020000	2.020000	2.060000	1.300000	2.300000	2.240000	2.220000	3.620000	3.200000	2.760000	3.160000	4.080000	2.860000	3.680000	4.300000	4.220000	3.400000	4.260000	3.960000	5.260000	4.840000	4.000000	5.100000	3.740000	5.380000	5.980000	5.720000	4.200000	4.380000	7.470000	6.650000	7.190000	8.190000	8.030000	6.830000	9.010000	5.780000	6.220000	8.810000	9.190000	10.350000	9.610000	10.510000	9.450000	11.030000	11.470000	12.590000	12.610000	0.000000	14.950000	19.440000	19.500000	81.930000
2.9 * IQR	0.000000	0.000000	0.300000	0.420000	0.400000	0.000000	1.300000	1.020000	1.720000	1.360000	0.700000	1.480000	1.520000	0.700000	1.740000	1.620000	1.580000	3.100000	2.600000	2.080000	2.560000	3.300000	2.100000	2.820000	3.440000	3.360000	2.740000	3.480000	2.780000	4.500000	3.880000	2.940000	4.200000	3.060000	4.300000	5.080000	5.080000	3.480000	3.320000	6.790000	5.400000	6.350000	7.630000	7.070000	5.960000	8.090000	4.060000	4.840000	7.870000	7.810000	9.410000	8.750000	9.310000	7.930000	9.750000	9.930000	11.050000	11.090000	0.000000	13.470000	19.260000	19.460000	78.340000
3 * IQR	0.000000	0.000000	0.280000	0.380000	0.360000	0.000000	1.280000	0.960000	1.680000	1.180000	0.660000	1.360000	1.480000	0.700000	1.660000	1.460000	1.480000	3.020000	2.480000	1.940000	2.460000	3.120000	2.040000	2.680000	3.240000	3.180000	2.640000	3.340000	2.640000	4.360000	3.700000	2.760000	3.980000	2.920000	4.080000	4.980000	4.920000	3.180000	3.100000	6.590000	5.220000	6.120000	7.510000	6.890000	5.680000	7.970000	3.860000	4.540000	7.610000	7.590000	9.190000	8.430000	9.050000	7.730000	9.330000	9.590000	10.570000	10.710000	0.000000	13.130000	19.240000	19.420000	77.560000

	f1_score	precision	recall	accuracy
2.000000	0.402299	0.269231	0.795455	0.919629
2.200000	0.420382	0.292035	0.750000	0.929675
2.300000	0.412500	0.284483	0.750000	0.927357
2.500000	0.394737	0.277778	0.681818	0.928903
3.300000	0.462810	0.363636	0.636364	0.949768
3.400000	0.525424	0.418919	0.704545	0.956723
3.700000	0.512397	0.402597	0.704545	0.954405
3.800000	0.529915	0.424658	0.704545	0.957496
3.900000	0.521739	0.422535	0.681818	0.957496

Problem Statement¶

Attribute Information¶

Importing the libraries¶

Data Preparation¶

Checking for categorical Variables¶

Checking for Missing values¶

Identifying numerical columns¶

Feature Elimination : from correlation (Pearson correlation)¶

We are not able to compare how the features are correlated to each other¶

Pearson correlation¶

Checkpoint 1¶

EDA + Outlier Analysis and Noise Removal and Visualization¶

Checking skewness of all the features in the form of distribution¶

Above we can see top 5 positive skewed features and top 5 negative skewed features¶

Lets start analyzing features that skewed from right to left¶

NOTE This is not any re-sampling technique as it has to be performed after splitting the dataset¶

Intension of this Analysis of skewness of features is to remove the noise at the end or beginning of the distribution Like values above .99 percentile and below .01 percentile that are very peculiar or uneven¶

The illustration of purpose of this intension can be seen below while removing noise from features that have skewness from range 70 to 85¶

Analysing Feature from skewness 70 to 85¶

Features Between the skewness range 70 to 85 :¶

Analysing Features having skewness from 60 to 70¶

Analysing Features having skewness from 45 to 61¶

Analysing Features having skewness from 30 to 44¶

Analysing Features having skewness from 20 to 30¶

Analysing Features having skewness from 13 to 20¶

Analysing Features having skewness from 7 to 10¶

Analysing Features having skewness from 4 to 7¶

Skewness of features from -1 to 4 is been ignored as they are less skewed and they might be handled later after splitting the dataset¶

Analysing Features having skewness from -10 to -2¶

Analysing Features having skewness from -40 to - 20¶

Analysing Features having skewness from -85 to -50¶

Checkpoint 2¶

Splitting data¶

NOTE!!! The variable [bank_data_stage_2] is a training set (X_train y_train) combined¶

1. Training Set:¶

2. Testing Set:¶

Hence the dataset is split we can perform any sampling method on training dataset¶

Feature Selection :Select the best features and rank them in order using¶

(Select KFold Technique)¶

We can use this ranked order features and sort them from high rank to rank or high score to low score and checking the model performance for each Addition of features starting from high to low¶

NOTE!!! This can be also used for tuning the model and checking the model performance¶

For Example:¶

1. Treating standard scale value IQR formula i.e 1.5 of as an hyperparameter while handling outliers by varying it from 1.5 ,1.6,1.7 and so on....¶

Problem: But!! dont know how much outliers on feature basis we cant eliminate¶

2. Solution:¶

And!! Now we got an another hyper parameter that is handling outlier for numbers for features, handling from feature that has less to outliers occurrence till the feature that has maximum outlier occurence¶

For We have two hyper parameter to tune¶

Only data point of majority class is been fed Because there is no compromise for modifying data of minority class¶

Suppose you want to select first 15 features for eliminating outliers with IQR scale range from (1.5 to 3.0)¶

We also have the list of columns with ranking of features which we have from SelectK technique¶

Models¶

Tuning data by IQR scale for XGBoost model¶

1. Tuning model by Feature Selection for XGBoost Model¶

Hypertuning XGBoost model with model parameters to check the improvement in performance¶

Tuning data for Random Forest model for IQR scale .¶

Tuning model by Feature Selection for Random Forest Model¶

Hypertuning Random Forest with model parameters to check the improvement in performance¶

Tuning data for GradientBoost model for IQR scale from 2.3 to 3.9¶

Tuning model by Feature Selection for Gradient Boost Model¶

Gradient boost Model have same performance after tuning with high ranked feature selection method¶

Hypertuning GradientBoost with model paramaters¶

Finally gathering the best parameter tuned and comparing the models performance¶

Models Used:¶

1. XGBoost (Best)¶

2. Random Forest (Best)¶

Random Forest Model Performace¶

ROC Curve of models¶

Precision Recall Curve (Percision Recall Trade Off)¶

	f1_score	precision	recall	accuracy
50	0.512000	0.395062	0.727273	0.952859
51	0.504065	0.392405	0.704545	0.952859
52	0.545455	0.428571	0.750000	0.957496
53	0.504202	0.400000	0.681818	0.954405
54	0.500000	0.402778	0.659091	0.955178
55	0.529915	0.424658	0.704545	0.957496

	f1_score	precision	recall	accuracy
2.100000	0.423077	0.294643	0.750000	0.930448
2.200000	0.438710	0.306306	0.772727	0.932767
2.300000	0.453333	0.320755	0.772727	0.936631
2.400000	0.463576	0.327103	0.795455	0.937403
2.500000	0.447552	0.323232	0.727273	0.938949
2.600000	0.455172	0.326733	0.750000	0.938949
2.700000	0.428571	0.312500	0.681818	0.938176
2.800000	0.421875	0.321429	0.613636	0.942813
2.900000	0.447761	0.333333	0.681818	0.942813
3.000000	0.449275	0.329787	0.704545	0.941267

	f1_score	precision	recall	accuracy
61	0.408163	0.291262	0.681818	0.932767
62	0.419580	0.303030	0.681818	0.935858
63	0.386667	0.273585	0.659091	0.928903
64	0.418919	0.298077	0.704545	0.933539
65	0.413793	0.297030	0.681818	0.934312
66	0.413793	0.297030	0.681818	0.934312
67	0.434211	0.305556	0.750000	0.933539
68	0.432432	0.307692	0.727273	0.935085
69	0.453333	0.320755	0.772727	0.936631
70	0.410959	0.294118	0.681818	0.933539
71	0.438356	0.313725	0.727273	0.936631
72	0.427586	0.306931	0.704545	0.935858
73	0.438356	0.313725	0.727273	0.936631

	f1_score	precision	recall	accuracy
2.300000	0.353488	0.222222	0.863636	0.892581
2.400000	0.371859	0.238710	0.840909	0.903400
2.500000	0.391960	0.251613	0.886364	0.906491
2.600000	0.371134	0.240000	0.818182	0.905719
2.800000	0.385027	0.251748	0.818182	0.911128
3.000000	0.365482	0.235294	0.818182	0.903400
3.100000	0.372340	0.243056	0.795455	0.908810
3.200000	0.358696	0.235714	0.750000	0.908810
3.300000	0.385027	0.251748	0.818182	0.911128
3.400000	0.376344	0.246479	0.795455	0.910355
3.500000	0.391304	0.257143	0.818182	0.913447
3.600000	0.400000	0.262411	0.840909	0.914219
3.700000	0.382979	0.250000	0.818182	0.910355
3.800000	0.400000	0.264706	0.818182	0.916538
3.900000	0.384615	0.253623	0.795455	0.913447
4.000000	0.393443	0.258993	0.818182	0.914219
4.100000	0.382514	0.251799	0.795455	0.912674
4.200000	0.345946	0.226950	0.727273	0.906491

	f1_score	precision	recall	accuracy
51	0.359447	0.225434	0.886364	0.892581
52	0.356808	0.224852	0.863636	0.894127
53	0.370732	0.236025	0.863636	0.900309
54	0.362745	0.231250	0.840909	0.899536
55	0.360976	0.229814	0.840909	0.898764
56	0.378109	0.242038	0.863636	0.903400
57	0.383420	0.248322	0.840909	0.908037
58	0.362637	0.239130	0.750000	0.910355
59	0.368715	0.244444	0.750000	0.912674
60	0.375691	0.248175	0.772727	0.912674
61	0.378378	0.248227	0.795455	0.911128
62	0.333333	0.218310	0.704545	0.904173
63	0.349206	0.227586	0.750000	0.904946
64	0.353591	0.233577	0.727273	0.909583
65	0.364641	0.240876	0.750000	0.911128
66	0.362637	0.239130	0.750000	0.910355
67	0.380435	0.250000	0.795455	0.911901
68	0.378378	0.248227	0.795455	0.911128
69	0.395604	0.260870	0.818182	0.914992
70	0.390805	0.261538	0.772727	0.918083
71	0.386740	0.255474	0.795455	0.914219
72	0.385027	0.251748	0.818182	0.911128
73	0.400000	0.262411	0.840909	0.914219